University of South FloridaScholar Commons

Graduate Theses and Dissertations Graduate School

7-4-2018

Development and Validation of the WaitingAssessment Interview Tool (WAIT) andEvaluation of Individualized Waiting Durations inSignaled ReinforcementClaudia Campos FleitasUniversity of South Florida, claudia24@mail.usf.edu

Follow this and additional works at: https://scholarcommons.usf.edu/etd

Part of the Behavioral Disciplines and Activities Commons, and the Social and BehavioralSciences Commons

This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion inGraduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contactscholarcommons@usf.edu.

Scholar Commons CitationCampos Fleitas, Claudia, "Development and Validation of the Waiting Assessment Interview Tool (WAIT) and Evaluation ofIndividualized Waiting Durations in Signaled Reinforcement" (2018). Graduate Theses and Dissertations.https://scholarcommons.usf.edu/etd/7270

Development and Validation of the Waiting Assessment Interview Tool (WAIT) and Evaluation

of Individualized Waiting Durations in Signaled Reinforcement

by

Claudia Campos Fleitas

A dissertation submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy Department of Child and Family Studies

College of Behavioral and Community Sciences University of South Florida

Major Professor: Sarah E. Bloom, Ph.D., BCBA-D Chair: Nicole Hanney, Ph.D., BCBA-D Catia Cividini-Motta, Ph.D., BCBA-D Andrew L. Samaha, Ph.D., BCBA-D

Eric Storch, Ph.D.

Date of Approval: June 29, 2018

Keywords: functional communication training, indirect assessments, problem behavior, signaled reinforcement, waiting skills

Copyright © 2018, Claudia Campos Fleitas

DEDICATION

This dissertation is dedicated to the strongest women in my life: Mom, Julie, and Sofie.

ACKNOWLEDGMENTS

This research would have not been possible without the mentorship of my advisor, Dr.

Sarah E. Bloom. I am grateful for all your professional and personal support and guidance.

Thank you for believing in me! In addition, I would like to thank Dr. Yanerys Leon and Dr.

Catia Cividini-Motta for shaping my skills as a researcher, teacher, and practitioner.

To the rest of my committee members, Dr. Andrew L. Samaha, Dr. Eric A. Storch, and

Dr. Nicole Hanney, I am thankful for the valuable feedback you have all provided me throughout

this process. I would also like to acknowledge Jennifer R. Weyman, Dr. Anna R. Garcia, Marissa

Lewis, Dr. Casey Clay, and Dr. Megan Boyle for their assistance with the completion of this

study and Engage Behavioral Health for supporting and allowing me to recruit participants from

their clinics. Furthermore, I am forever grateful to the McKnight Foundation for supporting my

research and funding my doctoral studies.

Finally, I thank my entire family for their unconditional love and endless support. I am

especially thankful to my mother, Ileana M. Fleitas, and my father, Enrique R. Campos, who

taught me to love education, follow my dreams, and work very hard to accomplish my goals. To

my sister, Julie, thank you for being my first mentor in life and my biggest cheerleader. To my

brother, Shawn, thank you for making me laugh and for always finding the beautiful things in

life. Lastly, to my husband, Greg, thank you for making my days a little brighter and easier,

supporting all my goals and ideas, and experiencing the ups and downs of life with me.

i

TABLE OF CONTENTS

List of Tables ............................................................................................................................. iii List of Figures ............................................................................................................................ iv Abstract ...................................................................................................................................... v Chapter One: Introduction .......................................................................................................... 1

Autism Spectrum Disorder and Problem Behavior ........................................................... 1 Functional Analysis ......................................................................................................... 2

Chapter Two: Treatments for Problem Behavior ......................................................................... 5 Functional Communication Training ............................................................................... 5 Delay Reinforcement ....................................................................................................... 7 Schedule Thinning........................................................................................................... 8 Stimulus Control ........................................................................................................... 10

Chapter Three: Waiting ............................................................................................................. 13 Chapter Four: Questionnaires and Caregiver Predictions ........................................................... 17 Chapter Five: Study I ................................................................................................................ 21

Purpose ......................................................................................................................... 21 Method .......................................................................................................................... 21

Development of the Waiting Assessment Interview Tool ................................... 21 Expert Review ................................................................................................... 22

Results .......................................................................................................................... 23 Discussion ..................................................................................................................... 24

Chapter Six: Study II ................................................................................................................ 26

Purpose ......................................................................................................................... 26 Phase I: Indirect Assessments ........................................................................................ 26

Method .............................................................................................................. 26 Subjects .................................................................................................. 26 Demographic Questionnaire ................................................................... 27 Functional Analysis Screening Tool ....................................................... 27 Waiting Assessment Interview Tool ....................................................... 27 Inter-Rater Reliability of the WAIT ........................................................ 27

ii

Results ............................................................................................................... 28 Discussion ......................................................................................................... 28

Phase II: Direct Assessments ......................................................................................... 30 Method .............................................................................................................. 30

Subjects and Settings .............................................................................. 30 Preference Assessment ........................................................................... 32 Response Measurement, Reliability, and Treatment Integrity ................. 32 Latency Functional Analysis .................................................................. 33

Results ............................................................................................................... 35 General Method ................................................................................................. 37

Comparison of Functions........................................................................ 37 Linear Regression................................................................................... 38

Results ............................................................................................................... 39 Comparison of Functions........................................................................ 39 Linear Regression................................................................................... 39

General Discussion ............................................................................................ 40 Chapter Seven: Study III ........................................................................................................... 41

Purpose ......................................................................................................................... 41 Method .......................................................................................................................... 41

Subjects and Setting ........................................................................................... 41 Response Measurement, Reliability, and Treatment Integrity ............................. 41 Design ............................................................................................................... 42 Baseline ............................................................................................................. 42 Functional Communication Training ................................................................. 42 Schedule Thinning ............................................................................................. 43

Results .......................................................................................................................... 46 Baseline ............................................................................................................. 46 Functional Communication Training .................................................................. 46 Schedule Thinning ............................................................................................. 47

Discussion ..................................................................................................................... 50 Chapter Eight: Conclusion ........................................................................................................ 54 References ................................................................................................................................ 69 Appendix A: Waiting Assessment Interview Tool (WAIT) – English ....................................... 78 Appendix B: Waiting Assessment Interview Tool (WAIT) – Spanish ....................................... 79 Appendix C: Demographic Questionnaire ................................................................................. 80 Appendix D: Functional Assessment Screening Tool (FAST) ................................................... 81 Appendix E: IRB Approval Letter ............................................................................................ 82

iii

LIST OF TABLES

Table 1: Inter-Rater Reliability of WAIT (caregivers) ........................................................... 57

Table 2: Inter-Rater Reliability of WAIT (behavior service providers) .................................. 58

Table 3: Interobserver Agreement and Treatment Integrity Latency Functional Analyses ...... 59

Table 4: Comparison of Functions for WAITs and Latency Functional Analyses................... 60

Table 5: Interobserver Agreement for Study III ..................................................................... 61

Table 6: Treatment Integrity for Study III .............................................................................. 62

iv

LIST OF FIGURES

Figure 1: Expert review responses to initial WAIT questions .................................................. 63

Figure 2: Latency functional analyses for six subjects ............................................................. 64

Figure 3: Latency functional analyses for five subjects ........................................................... 65

Figure 4: Linear regression for function(s) of problem behavior ............................................. 66

Figure 5: Rate of problem behavior for four subjects during baseline, functional communication training, and multiple schedules. ..................................................... 67 Figure 6: Rate of functional communication responses for all four subjects during baseline, functional communication training, and multiple schedules. ...................... 68

v

ABSTRACT

Indirect assessments are widely used to identify environmental factors that may be

manipulated or integrated in the development of direct assessments and behavior interventions

for individuals with developmental disabilities. To date, there are no indirect assessments that

can be used to evaluate the extent to which children who engage in problem behavior wait to

receive reinforcement. However, there are effective behavior interventions to increase waiting. A

treatment package consisting of functional communication training (FCT) and schedule thinning

(i.e., multiple schedules) has been identified as an effective intervention to increase waiting in

children who engage in problem behavior maintained by social contingencies. Nonetheless, in

studies in which a schedule thinning procedure has been used, the terminal waiting durations are

typically selected arbitrarily. Therefore, we conducted three studies to evaluate the use of an

indirect assessment to increase waiting within an FCT and schedule treatment package in

children with developmental disabilities. The purpose of Study I was to develop the Waiting

Assessment Interview Tool (WAIT) to obtain current waiting durations for subjects who engaged

in problem behavior maintained by social contingencies. The purpose of Study II was to

complete the WAIT with caregivers and behavior service providers and to compare their results

to a latency functional analysis (FA) conducted with all children. Finally, the purpose of Study

III was to use the WAIT completed by informants to systematically individualize the initial

component durations used during the schedule thinning procedure. A second purpose was to use

the informants’ preferred waiting times as final waiting targets for all subjects.

1

CHAPTER ONE

INTRODUCTION

Autism Spectrum Disorder and Problem Behavior

It has been estimated that 15% of children ages 3 to 17 years old have been diagnosed

with at least one developmental disability in the United States (Centers for Disease Control and

Prevention; CDC, 2016). According to the CDC, developmental disabilities consist of a group of

disorders that impair physical and cognitive learning. These disorders or conditions are typically

diagnosed during the early stages of life and may drastically impact the daily activities of

individuals. One of the most prevalent developmental disabilities observed in children is autism

spectrum disorder (ASD).

The prevalence of children with ASD has drastically increased in recent decades. Perez,

Sawmiller, and Tan (2016) found that ASD is the second largest developmental disability in the

United States with 1 in every 68 children (i.e., 1 in every 48 boys, 1 in every 189 girls) being

diagnosed (Christensen et al., 2016). The Diagnostic and Statistical Manual of Mental Disorders

(DSM-V-TR; American Psychiatric Association, 2013) defines ASD as a neurodevelopmental

disorder that typically manifests during the early development of children. ASD significantly

impairs social communication and interaction, acquisition of skills, social reciprocity, and non-

verbal communication behavior. To be diagnosed with ASD, a social communication deficit and

restricted or repetitive, or both, behavior must significantly impact the daily life of the

individual. Moreover, there are several atypical behaviors that are frequently observed in

2

individuals with ASD (Dominick, Davis, Lainhart, Tager-Flusberg, & Folstein, 2007).

Dominick et al. (2007) identified the following problem behaviors as the most common: a)

feeding disorders (e.g., food selectivity), b) abnormal sleep patterns, c) self-injurious behavior, d)

aggression, and f) tantrums.

Although there are no known cures for ASD, there are research-based interventions (e.g.,

early intensive behavioral intervention; Lovaas, 1993) that may help increase the quality of life

and independent functioning of individuals with this disorder (Strain, Schwartz, & Barton, 2011).

These interventions typically include increasing appropriate skills (e.g., daily living activities,

verbal communication, social skills) and decreasing problem behavior. Functional analyses

(Iwata, Dorsey, Slifer, Bauman, & Richman, 1982/1994) have been evaluated to assess problem

behavior emitted by individuals with ASD. Also, some refinements have been made to the

traditional functional analysis (FA) to increase the practicality of the assessment for different

severities of responses, topography of the responses, and settings. These modified assessments

continue to be evaluated and refined. Their results are used to identify function-based treatments

for problem behavior.

Functional Analysis

The FA approach as conducted by Iwata et al. (1982/1994) is a direct assessment of

problem behavior in which antecedents and consequences are manipulated to identify the

environmental contingencies maintaining the problem behavior. Functional analyses are highly

recommended for practitioners and researchers to conduct when working with individuals who

engage in problem behavior (Carr & Fox, 2009). Functional analyses typically consist of several

test conditions and a control condition. All test conditions are compared to the control condition

and higher responses of problem behavior in one or more test conditions typically suggest that

3

problem behavior is maintained by the contingency in that condition. The typical conditions in a

FA are a) alone or ignore, b) attention, c) play, and d) escape. Sometimes access to tangible

items is evaluated as well.

As previously mentioned, researchers have continued to evaluate the FA to maximize its

benefits while reducing some potential risks for subjects and implementers. Beavers, Iwata, and

Lerman (2013) identified more than 2000 published studies and chapters have extended and

replicated functional analyses. Replications have included different topographies of problem

behavior, subjects, settings, and procedures. Several variations of functional analyses have also

addressed some of its potential limitations. For example, the brief FA (Northup et al., 1991) was

developed to address the total duration of the assessment. The precursor FA (Smith & Churchill,

2002) evaluated the use of precursor responses for severe problem behavior to decrease the risks

associated with conducting a FA for severe topographies. The extended alone condition

(Querium et al., 2013) assessed the use of alone conditions prior to all other conditions to

identify potential automatically-maintained functions of problem behavior to increase the

efficiency of the assessment by not having to conduct all conditions of the FA. Trial-based FA

(Bloom, Iwata, Fritz, Roscoe, & Carreau, 2011) was developed to increase the practicality of the

assessment by conducting it less controlled settings (e.g. educational settings). Finally, the

latency FA (Thomason-Sassi, Iwata, Neidert, & Roscoe, 2011) evaluates the use of latency as a

measure of behavior rather than the rate at which problem behavior occurs. Latency is the

interval of time between the offset of a stimulus to the onset of a response (Skinner, 1938). In

most traditional functional analyses and variations of it, the dependent variable is the repetition

of behavior (Thomason-Sassi et al., 2011). However, when problem behavior is very severe it

may not be appropriate to allow the individual to engage in multiple instances of the response.

4

Therefore, if latency data are collected, it is possible to evaluate the contingencies that maintain

problem behavior by analyzing the latency of engaging in the problem behavior in each

condition. Moreover, when the occurrences of a behavior terminate a session due to the

topography of the response (e.g., elopement), a latency FA could be used as the method of

assessment.

Latency functional analyses consist of the same conditions as the conventional FA. The

dependent measure is the time from the onset of a stimulus to the first instance of the response

(Thomason-Sassi et al., 2011). Sessions are conducted in a multielement design using the

following sequence: a) alone, b) attention, c) play, and d) escape. Each session ends after a single

instance of the problem behavior (and the immediately following delivery of the programmed

consequence) or after five min have elapsed. If problem behavior occurs during the alone or play

conditions, sessions are terminated after 1 min of the problem behavior to avoid advantageous

reinforcement.

Based on the results of the functional analyses, several function-based treatments have

been evaluated and found to be successful. Functional communication training (FCT; Carr &

Durand, 1985) is one of the most commonly used treatments when results from a FA identify a

social reinforcement contingency (e.g., positive reinforcement in the form of attention, negative

reinforcement in the form of escape from demands; Hagopian, Boelter, & Jarmolowicz, 2011).

5

CHAPTER TWO

TREATMENTS FOR PROBLEM BEHAVIOR

Functional Communication Training

Functional communication training is a differential reinforcement procedure in which the

individual who engages in problem behavior is taught a functional communication response

(FCR) to request for the reinforcer (Carr & Durand, 1985). This response should be a functional

mand that is under the control of the same establishing operation (EO; Laraway, Snycerski,

Michael, & Poling, 2003) as the problem behavior. During FCT, the FCR is reinforced on a

dense schedule whereas the problem behavior is placed on extinction. Using extinction while

implementing FCT facilitates problem behavior reduction (Fisher, Thompson, Hagopian,

Bowman, & Krug, 2000) and the acquisition of the new communication response. Functional

communication training has been widely researched in the literature and has found to be a

successful intervention for socially mediated problem behavior (Hagopian et al., 2011).

Furthermore, Tiger, Hanley, and Bruzek (2008) identified FCT has the most published treatment

for problem behavior maintained by social contingencies. However, a limitation of the

intervention is the impracticality of reinforcing the communication response when it occurs at

very high rates or during situations in which reinforcement is not available (Fisher et al., 2000).

Generalization and maintenance of the FCR could be difficult to obtain if the reinforcement

schedule is not thinned as part of the intervention. Furthermore, previously reinforced problem

behavior could reemerge to baseline levels of responding if the FCR is not reinforced

consistently across settings and caregivers (Tiger et al., 2008).

6

Tiger et al. (2008) suggested some practical guidelines to promote the generalization and

maintenance of FCRs in natural environments. Some of these guidelines include the

manipulation of response effort (Horner & Day, 1991), social recognition of the FCR, extinction

of problem behavior, time delays between responses and delivery of reinforcers, reinforcement

thinning, and the use of stimulus control procedures. To increase the likelihood of an appropriate

communication response, the communication response’s effort should be equal to, or preferably

less than, the response effort for the problem behavior. Meaning, engaging in the communication

response should be easier for the individual than engaging in the problem behavior. If the

communication response is more effortful (e.g., a vocal response for a non-vocal individual) than

the problem behavior, the communication response is less likely to be emitted. Thus, the new

contingency may be difficult to establish. Moreover, if the FCR is chosen arbitrarily (e.g., sign

not recognized in sign language), the individual may be placed in settings in which others may

not know how to reinforce the response, which may weaken the response-reinforcer contingency

and may result in the reemergence of problem behavior. Another very important guideline when

implementing FCT is the use of extinction, or a thinner schedule of reinforcement, for the

problem behavior. If engaging in the problem behavior continues to be reinforced on the same

schedule as the new FCR, the individual is more likely to engage in the problem behavior due to

the previous history of reinforcement with this response versus the new one.

The last three recommendations involve the implementation of delayed reinforcement,

schedule thinning, and using stimulus control procedures to address the limitation of FCT. This

limitation occurs when individuals engage in the communication response at rates that cannot be

reinforced by the others in their environment (e.g., ask for attention every 5 s or for a break from

academic work every 30 s) or when reinforcement cannot be provided (e.g., iPadÒ does not have

7

battery).

Delayed Reinforcement

Response chaining is one procedure used to address the limitation of FCT. Lalli, Casey,

and Kates (1995) used response chaining with three individuals who engaged in problem

behavior maintained by negative reinforcement. Lalli et al. taught the individuals communicative

responses to decrease problem behavior and implemented a response chaining procedure to teach

the individuals to tolerate delays to reinforcement. The response chaining procedure consisted of

gradual increases in the response requirement before presenting the opportunity to request for a

break from demands. The results from this study suggest that using response chaining was very

effective in decreasing problem behavior and increasing the time between the FCR and the

reinforcer (i.e., break). Similarly, Fisher et al. (1993) evaluated the use of this procedure for

behavior maintained by positive reinforcement by increasing the waiting time between the

response and the delivery of the reinforcer (e.g., attention, tangibles). Delayed reinforcement for

behavior maintained by positive reinforcement was also effective in addressing the limitation of

FCT. Reinforcer delay fading has also been used to teach children how to engage in “self-

control”, by teaching them how to choose a larger later reward instead of an immediate smaller

reward (Schweitzer & Sulzer-Azaroff, 1988).

An alternative procedure that has been evaluated in the literature to increase tolerance to

delayed reinforcement is the use of another activity or reinforcers during the waiting period.

Campos, Leon, Sleiman, and Urcuyo (2016) provided access to a positive reinforcer (i.e., toy,

food) contingent on compliance to tasks during an extinction component of a multiple schedule

to decrease FCRs for breaks when reinforcement was not available. This procedure was effective

for one of the two subjects in the study. The results from this study suggest that when individuals

8

engage in a FCR and the reinforcer is not available, it might be useful to present another activity

for which another form of reinforcement can be provided. Having a competing activity may

decrease the establishing operation to request for the reinforcer and engage in problem behavior

while waiting.

Learning to wait is an important skill for all individuals, including children with

developmental disabilities (Fisher et al., 2000). As previously mentioned, there are situations in

which receiving the requested reinforcer might not be possible or practical. Therefore, teaching

individuals to wait should be embedded in our clinical practice, especially when teaching a FCR

to decrease problem behavior in individuals with developmental disabilities. Consequently,

developing and evaluating procedures that could reduce the reemergence of problem behavior

and the requests for functional reinforcers during periods of nonreinforcement continues to be an

important topic of research.

Schedule Thinning

Hanley, Iwata, and Thompson (2001) evaluated the use of four reinforcement schedule

thinning procedures after implementing FCT for individuals with developmental disabilities who

engaged in problem behavior maintained by positive reinforcement. Three subjects learned how

to request for the functional reinforcer using FCT. One of the subjects was then exposed to three

reinforcement schedule-thinning procedures; increasing delays to reinforcement, graduated fixed

interval (FI) schedule, and multiple schedule. The other two subjects were then exposed to mixed

and multiple schedules of reinforcement.

In Hanley et al. (2001), the increasing delays to reinforcement schedule consisted of

gradually increasing delays between the FCR and the reinforcer delivery. The maximum delay

reached was 25 s. However, the FCR was extinguished during this schedule. In the graduated FI

9

schedule, the functional reinforcer was delivered after the first FCR was emitted after the current

interval had elapsed. The final FI was 58 s. This schedule of reinforcement produced very high

rates of the alternative response, which made it impractical for individuals in the natural

environment to reinforce.

After evaluating the increasing delays and graduate FI schedules, a multiple schedule was

implemented for the first subject. The multiple schedule consisted of two components that

alternated. One component was reinforcement on a fixed ratio 1 (FR 1) and the other was

extinction. Both components were signaled with corresponding stimuli (i.e., white and red

cards). During the reinforcement component, all FCRs received reinforcement. During the

extinction component, all FCRs were placed on extinction. Problem behavior was placed on

extinction during both components. The component durations increased gradually after two

consecutive sessions in which problem behavior was less than, or equal to, 85% reduction from

baseline. The initial component durations were 45-s reinforcement / 15-s extinction and the final

component durations were 60-s reinforcement / 240-s extinction. In a 10-min session, there were

two 1-min reinforcement components and two 4-min extinction components. The multiple

schedule evaluated produced stable rates of the communication response during the

reinforcement component and almost zero rates during the extinction component. Furthermore,

problem behavior remained near zero rates during both components.

Because of the successful results of the multiple schedule with the first subject, the

authors decided to evaluate the same procedure with subjects two and three. Moreover, to assess

the role of the schedule-correlated signals, they compared the multiple schedules to mixed

schedules of reinforcement for the two subjects. The multiple schedules evaluated for these

subjects were identical to the one used for subject one. Furthermore, the mixed schedule was the

10

same as the multiple schedule with one exception, no schedule-correlated signals were present.

The results from these two subjects suggested that the use of schedule-correlated signals allowed

for the discrimination of the components and facilitated discriminated manding. Meaning, using

schedule-correlated signals facilitated schedule thinning.

Stimulus Control

Hanley et al. (2001) were successful in thinning the schedule of reinforcement for three

individuals with developmental disabilities by using multiple schedules. Their results have been

replicated and expanded in the literature. For example, Betz, Fisher, Roane, Mintz, and Owen

(2013) used multiple schedules with four individuals with problem behavior maintained by

positive reinforcement. Betz et al. taught the subjects FCRs before evaluating mixed and

multiple schedules of reinforcement with rapid alternation of components. The rapid alternation

consisted of changes between the reinforcement and extinction components. Both components

were 60 s in duration and no gradual schedule thinning was evaluated. After establishing

discriminated manding with the 60-s reinforcement / 60-s extinction, the authors increased the

extinction component to 240 s. Similarly to Hanley et al. (2001), in a 10-min session, there were

two 4-min extinction components and two 1-min reinforcement components. Furthermore, the

authors provided specific rules during both, multiple and mixed schedules. The results from the

study suggest that specific rules facilitated discriminated manding only when schedule-correlated

stimuli were present. When the rules were provided during the mixed schedule, in which

schedule-correlated stimuli were not present, the rules had little or no effect on responding.

Therefore, discriminated manding was facilitated by the stimulus control in the multiple

schedules. Moreover, all subjects showed rapid acquisition of discriminated manding while

problem behavior remained low or near zero rates. Tiger and Hanley (2004) also assessed the use

11

of mixed and multiple schedules while providing rules in a classroom setting. In this study, three

typically developing children learned how to request for attention from their teacher. Children

were trained at the terminal schedule of reinforcement without the use of gradual fading.

In another evaluation of multiple schedules, Kuhn, Chirighin, and Zelenka (2010) taught

two individuals who engaged in problem behavior to request for reinforcers using functional

communication. Then, they used natural stimuli rather than arbitrary cards to signal the

reinforcement and extinction components. The natural signals consisted of therapists engaging in

busy (e.g., cooking) and non-busy (e.g., watching TV) activities that individuals would typically

encounter in their natural environment. Both subjects in this study learned to engage in

discriminated manding based on the overt behavior of the therapists engaging in the busy and

non-busy activities. In an extension, Leon, Hausman, Kahng, and Becraft (2010) replicated Kuhn

et al. (2010) across different settings, therapists, and novel sites, to increase the generality of

treatment effects. Leon et al. (2010) taught a young boy with developmental disabilities to

request for attention by saying “excuse me” during busy and non-busy times. Busy times

represented the extinction component and non-busy times represented the reinforcement

component.

Other studies have continued to evaluate the use of multiple schedules as a schedule

thinning procedure after implementing FCT. However, no research to date has evaluated a

systematic approach to identify appropriate individualized component durations during multiple

schedules. Most studies using multiple schedules have increased the components to arbitrary

durations such as 60-s reinforcement / 240-s extinction (e.g., Hanley et al., 2001; Betz et al.,

2013), 60-s reinforcement / 300-s extinction (e.g., Fisher, Greer, Fuhrman, & Querim, 2015), 60-

s reinforcement / 540-s extinction (e.g., Campos et al., 2016) and 300-s reinforcement / 900-s

12

extinction for one subject (Rooker, Jessel, Kurtz, & Hagopian, 2013). Therefore, it is unknown

whether individuals’ waiting histories could be used as the initial extinction durations and

terminal goals in the total duration of components in multiple schedules.

13

CHAPTER THREE

Waiting

Waiting, or tolerating delays, is defined as the latency between an opportunity to engage

in a response and the individual engaging in that response. For example, if a child has the

opportunity to ask for a cookie now, but asks for a cookie in 10 min, the child has waited for 10

min. Tolerating delays without engaging in problem behavior is considered a preschool life skill

(Hanley, Heal, Tiger, & Ingvarsson, 2007).

Individuals who engage in impulsive behavior, those who respond for a sooner smaller

reward instead of waiting for a larger later reward, are said to have less self-control (Vollmer et

al., 1999). Individuals typically encounter concurrent schedules in which they have to make

choices to respond on different schedules of reinforcement. Thus, individuals who successfully

wait by consistently selecting the larger later reward relative to the sooner smaller reward are

considered less impulsive. Individuals with developmental delays who have limited

communication repertoires typically lack waiting skills and may engage in more impulsive

behavior (Vollmer et al., 1999).

Previous research has used different procedures to teach individuals to make less

impulsive behaviors (e.g., waiting for longer periods of time to receive larger reinforcers). Some

of these procedures include delay fading (e.g., Vollmer et al., 1999), delay fading with

alternative activities (e.g., Dixon & Cummings, 2001), teaching rules to be repeated while

waiting (e.g., Hanley et al., 2007), using preferred toys (e.g., Newquist, Dozier, & Neidert,

2012), and providing qualitative different reinforcers (e.g., Passage, Tincani, & Hantula, 2012).

14

Vollmer et al. (1999) assessed self-control and impulsivity in two individuals with

developmental disabilities who engaged in aggression maintained by access to food, TV, or both.

The authors also evaluated the effects of signaled delays on impulsive aggression. Initially, the

children were taught to request for the reinforcers using FCT with and without a delay. Then, the

authors conducted reinforcer magnitude and impulsivity tests. The reinforcer magnitude tests

consisted of providing a greater amount of the reinforcer for the alternative response and a

smaller amount of the same reinforcer for aggression during a 0-s delay. The results from this

test demonstrated that at a 0-s delay, rate of responding was higher for the greater amount of

reinforcers than for the smaller amount of reinforcers. Hence, appropriate mands for reinforcers

occurred more often than the problem behavior. Moreover, when delays to reinforcement were

gradually increased in the impulsivity tests, children had to wait for the reinforcers for longer

durations. Here, both subjects demonstrated self-control by engaging in the appropriate behavior

only when signals for the delays were provided.

In another study, Dixon and Cummings (2001) evaluated self-control in children with

ASD by examining their response allocation. During baseline, the children were asked to wait for

as long as they could before eating or playing with their preferred items. The items were

provided when the children requested them or said that they could not wait any longer. The

children were then exposed to two choice baselines. In the first one, they were asked to select

between a small immediate reinforcer and a large delayed reinforcer. In the second one, they

were asked to select between an immediate small reinforcer and an immediate large reinforcer.

Finally, the children were exposed to a self-control procedure in which they had to choose

among a small immediate reinforcer, a large delayed reinforcer without a response requirement,

and a large delayed reinforcer with a response requirement. In this study, all children preferred

15

the small immediate reinforcer during the first baseline, the large immediate reinforcer during the

second baseline, and the large reinforcer with an activity during the self-control phase. Problem

behavior was near zero for all children during the self-control intervention that consisted of the

large reinforcer with activity. These results suggest that self-control may be increased in

individuals with disabilities by gradually exposing them to small increments in delays to receive

access to larger reinforcers. Also, this study demonstrated that providing choices to complete

activities during delays to reinforcement might minimize problem behavior and help increase

waiting.

Hanley et al. (2007) taught a group of students ages 3 to 5 years old to engage in a

mediating response that consisted of repeating rules (i.e., “When I wait quietly, I get what I

want”) while they waited for their reinforcers in a classroom setting. This procedure was used to

increase waiting in preschoolers. Some of the students had a developmental disability diagnosis

and some of the students were typically developing. In this study, repeating rules increased

waiting during delays and decreased problem behavior emitted by the children.

The combination of delay fading with different procedures (e.g., timers, rules, activities)

has been found to increase waiting. However, the extent to which these different procedures are

effective in the absence of delay fading has not received as much attention in the literature. A

noteworthy exception is a study conducted by Newquist et al. (2012) that evaluated the effects of

rules, timers, and the use of preferred toys without the use of delay fading with three typically

developing preschool children. The results from this study were slightly inconsistent with prior

research. In this study, providing rules and countdown timers during the delay was ineffective at

teaching waiting, however, providing toys during the delay was effective at increasing waiting.

In summary, different procedures have been evaluated to teach children and individuals

16

with developmental disabilities to wait. This is because waiting is considered an important life

skill. Waiting may result in increased access to preferred activities, items, and social interactions

(Newquist et al., 2012). Typically, individuals encounter several situations in which they are

required to wait since very early in life (e.g., waiting for parents to attend, waiting for their turn

in school). These situations may evoke problem behavior when the delays to reinforcement are

longer than they have previously encountered (e.g., history with delays to reinforcement) or

when waiting skills are not in their repertoire. Moreover, teaching children and individuals with

developmental disabilities to wait is important because this skill is a prerequisite for more

difficult daily living and academic activities (Newquist et al., 2012). Therefore, developing

procedures that increase the feasibility for practitioners to teach waiting skills in children with

ASD is an important component of treatment interventions.

17

CHAPTER FOUR

Questionnaires and Caregiver Predictions

Questionnaires, interviews, checklists, and rating scales are typically completed when

researchers and practitioners are interested in obtaining information about the variables that

influence human behavior and selections. In the field of behavior analysis, when individuals

engage in problem behavior that warrants assessment and treatment, questionnaires and

interviews identify useful information that could be further analyzed through direct assessments

(e.g., FA). Questionnaires could also be used when the interest is to identify potential sources of

reinforcement (e.g., type of attention, variety of foods and toys). Typically, subjects, parents,

caregivers, or teachers, or a combination of these, complete the questionnaires. This is because

these individuals have a longer history with the subjects than the researchers and practitioners. In

many cases, the information gathered from these indirect assessments is very valuable for the

development of interventions.

The Motivation Assessment Scale (MAS; Durand & Crimmins, 1988) and the Questions

about Behavior Function (QABF; Matson & Vollmer, 1995) were initially developed as tools to

identify potential sources of reinforcement for problem behavior in individuals with disabilities.

However, Zarcone, Rodgers, Iwata, Rourke, and Dorsey (1991) raised questions about the

reliability of these indirect assessments and cautioned not to use them in isolation to determine

the function of a behavior.

The Functional Analysis Screening Tool (FAST; Iwata, DeLeon, & Roscoe, 2013) was

developed to have an organized interview as a preliminary step to completing an FA. The FAST

18

was structured for the informants (e.g., parents, staff, teachers) to answer questions related to the

specific environmental contingencies under which problem behavior is most likely to occur. The

questions are divided by the potential source of reinforcement that can be maintaining the

problem behavior. After the development of the initial questions for the FAST, the questions

were used for initial assessments of problem behavior of individuals with developmental

disabilities at a residential facility. The individuals using these questions provided written

feedback about content and format. The reliability of the revised FAST was then evaluated in

three different facilities by administering the questions to pairs of direct-care staff who worked

with the individuals who engaged in the problem behavior. Following this step, items with low

scores were deleted or revised. A second revision of the FAST was then administered to a second

group of pairs of informants who worked with the individuals who engaged in the problem

behavior for an interrater reliability analysis. A validity analysis was completed after the

reliability analysis. The validity analysis consisted of comparing the function identified on each

pair of the FAST to the function identified by completing an FA. Iwata et al. (2013) used values

of 0, .5, or 1 to identify if both informants’ results corresponded with the function identified by

the FA (1), only one of the informants’ results corresponded with the function identified by the

FA (.5), or neither of the informants’ results corresponded with the function identified by the FA

(0).

The results from Iwata et al. (2013) identified what the authors had initially anticipated

based on the data from previous studies (e.g., Zarcone et al., 1991), low reliability and low

validity of the FAST. Due to the nature of the data collected by indirect assessments, it is very

difficult to have reliable and valid results. Data have suggested that parents and caregivers are

biased and make errors when answering these types of assessments (Iwata et al., 2013).

19

Moreover, Green, Reid, Canipe, and Gardner (1991) identified that parents and caregivers’

results have poor correspondence with direct assessments when it comes to subjects’ preferences

and selections.

Green et al. (1991) extended previous research on reinforcement and preference

assessments by completing stimulus preferences using behavioral assessments with individuals

with developmental disabilities and comparing their results to the selections made by their

caregivers. Their results suggested that the opinions of caregivers are not predictions of the

stimulus preferences of individuals with developmental disabilities as identified by direct

measurements.

Nevertheless, the use of these indirect assessments is still beneficial because they guide

and facilitate further assessment and treatment. Thus, practitioners and researchers continue to

use them. Moreover, indirect assessments have been expanded to other applications in ABA.

Some other indirect assessments used in the field include the Reinforcer Assessment for

Individuals with Severe Disabilities (RAISD; Fisher, Piazza, Bowman, & Amari, 1996) used to

identify potential stimuli that may serve as reinforcers for skill acquisition programs, Sleep

Assessment and Treatment Tool (Jin, Hanley, & Beaulieu, 2013) used to identify environmental

factors that may influence sleep patterns, and Matson Evaluation of Social Skills with

Youngsters (MESSY; Matson, Rotatori, & Helsel, 1983) and Social Skills Improvement System

(SSIS; Gresham & Elliott, 2008) used to identify social skills in young children.

Overall, indirect assessments continue to be widely used in the field of behavior analysis

possibly due to the fast acquisition of data that can be used for the development of direct

behavior assessments and interventions. Therefore, the purpose of the current set of studies was

to develop, implement, and validate a Waiting Assessment Interview Tool (WAIT) to obtain

20

current waiting durations for subjects who engage in problem behavior maintained by social

contingencies. A second purpose was to systematically individualize the component durations

used during multiple schedules and the final target durations for increasing waiting for each

subject.

21

CHAPTER FIVE

STUDY I

Purpose

The purpose of Study I was to develop the Waiting Assessment Interview Tool (WAIT)

to obtain current waiting durations for subjects who engaged in problem behavior maintained by

social contingencies.

Method

Development of the Waiting Assessment Interview Tool. A Waiting Assessment

Interview Tool (WAIT; see Appendix A for English version and Appendix B for Spanish

version) was developed for this study to identify the typical contexts and ranges of durations that

children are expected to wait to obtain access to preferred activities, attention, and to comply

with non-preferred tasks. The purpose of the WAIT was to identify current waiting performances

and potential individualized final target durations to be used to thin the schedule of

reinforcement in a way that is feasible for parents. A second purpose of the WAIT was to

determine gradual increments to thin the schedule of reinforcement during the extinction

components of a multiple schedule used for the subjects in study III.

The WAIT consists of four parts. The first part focuses on the demographic information

of the child and the caregiver completing the questionnaire. The target problem behavior is also

identified and operationally defined in this section. In addition, specific instructions on how to

fill out the questionnaire are provided.

The second part of the WAIT is divided into three subsections. These sections each

22

emphasize one of the three socially-maintained functions of problem behavior: 1) attention, 2)

tangible, and 3) escape. In each of these sub-sections, there are two questions that address how

long a child may wait to receive reinforcement before engaging in problem behavior and

caregivers have the option to provide additional information regarding the contexts in which

problem behavior happens. Caregivers also have the option to select non-applicable (N/A) if

their children’s problem behavior does not occur in the context asked in the questions. For

example, if the caregiver believes that problem behavior does not occur when the child is asked

to complete an activity, the caregiver may select N/A in the questions in found under the escape

section.

The third part of the WAIT is used to identify goals for appropriate individualized

waiting times for each child based on the caregivers’ preferences. The caregivers are asked to

select how long they would like their child to wait under different circumstances. The caregivers

are given the option to select a number from 5 to 60 and a unit of time (i.e., seconds or minutes).

Finally, the fourth part of the WAIT is intended for behavior analysts to analyze the

answers provided by the caregivers and identify the context (i.e., to get attention, access to

tangibles, or a break from work) in which problem behavior has the shortest latency. Meaning,

the child waits the shortest duration before engaging in problem behavior. Moreover, the

behavior analysts may also select the average current waiting duration by using the answers

provided by the caregivers, the initial treatment duration they would like to target, and the final

treatment duration they would like for the child to achieve.

Expert Review. Prior to developing the final WAIT, a 15-question draft of the

23

questionnaire was submitted to an expert review panel. This draft was completed using behavior

analytic literature on waiting and problem behavior. The questionnaire was provided through an

email that allowed the experts to access a link to the questionnaire and rate each question

anonymously using a 4-point rating scale as used by Dehdari, Rahimi, Aryaeian, Gohari, and

Esfeh (2014). The scale consisted of: not relevant (N), slightly relevant (S), relevant (R), and

very relevant (V). The experts also had the opportunity to provide comments for each question.

In addition, the experts were asked to rate the overall clarity of instructions, order of questions,

use of scale in every question, scoring summary, easy of completion, and overall format of the

questionnaire using a scale from zero (do not like) to ten (like).

The questionnaire was submitted to 21 Board Certified Behavior Analysts-Doctoral level

(BCBA-Ds) who had over five years of experience working with children with ASD who engage

in problem behavior when waiting, have conducted research on waiting or delayed

reinforcement, are authors of published questionnaires in the field of Applied Behavior Analysis

(ABA), or some combination thereof. Of the 21 BCBA-Ds who received the questionnaire, 7

BCBA-Ds provided feedback to all the questions.

Results

Of the initial 15 questions, the experts rated 93% of the questions as either relevant or

very relevant. All remaining questions were rated as slightly relevant and were removed from the

questionnaire (See Figure 1). In addition, the experts provided some comments regarding the

questions that were rated as relevant or very relevant. These comments suggested that some of

the information overlapped across questions and some of the questions were incomplete. To

address for overlap across questions, some questions were combined into one question. To

address for the incomplete wording of the questions, some questions were rewritten to add the

24

suggested feedback. For example, if the question was originally “How long would you like the

child to wait before you have to pay attention to him/her?” and the experts suggested to add

“without engaging in problem behavior,” the question was rewritten to “If the child wants your

attention and you are engaging in an activity (e.g., making dinner, driving, grading) and not

paying attention to the child, after how many minutes will the child engage in problem behavior

if you do not provide the attention requested?” Therefore, when questions rated slightly relevant

were removed and all comments about all other questions were addressed, 60% or 9 out of the

original 15 questions, became part of the final draft of the WAIT.

In addition, when asked about the overall questionnaire using a sliding scale from 0 to 10,

the expert panel rated the following: instructions (M = 7.14), order of questions (M = 8.00), use

of scale in every question (M = 8.29), scoring summary (M = 8.14), ease of questionnaire

completion (M = 8.14), overall format of the questionnaire (M = 8.00). To address this feedback,

the instructions were written more specifically (e.g., provided an example of how to answer the

questions), and the scoring summary and format of the questionnaire were changed. For

example, the final draft of the WAIT includes benchmarks for the parents to guide their answers

when asked about waiting goals.

Discussion

Study I resulted in the development of a questionnaire that potentially identifies how long

children wait before engaging in problem behavior under different circumstances. Furthermore,

the questionnaire asks the caregivers for the durations they would want their children to wait

before receiving reinforcement. This is important because including the caregivers’ preferences

and opinions may increase their buy-in and the integrity in which they implement treatment

procedures. Reimers, Wacker, and Koeppl (1987) suggested that the social acceptability of

25

treatments is necessary for generalization. Moreover, generalization is essential in the long-

lasting effects of treatments (Stokes & Baer, 1977). In addition, the WAIT may also be used by

behavior analysts to individualize waiting duration times in different settings as part of their

treatment packages for problem behavior.

26

CHAPTER SIX

Study II

Purpose

The purpose of Study II was to complete the WAIT with caregivers and behavior service

providers and to compare their results to a latency FA conducted with all children.

Phase I – Indirect Assessments

Method.

Subjects. The 33 subjects for Study II Phase I consisted of informants who completed the

WAIT for every child who participated in the study (see children characteristics in Phase II

method). Two caregivers (e.g., parents, siblings) and two behavior service providers (e.g., Board

Certified Behavior Analysts [BCBAs], Board Certified Assistant Behavior Analysts [BCaBAs],

Registered Behavior Technicians [RBTs], behavior therapists) were asked to serve as informants

for all children. This sample was uncontrolled but is representative of the natural sample that

would complete the WAIT if practitioners and researchers used this tool. Moreover, if children

did not receive ABA services, no data were available for the two behavior service providers

(David, Allen, Malik). In addition, if only one caregiver was available to complete the WAIT, no

data were available for a second caregiver (Blake, Alex, Javier, Allen, Malik). Overall, the

informants who completed the WAIT consisted of 11 mothers, 5 fathers, 1 male sibling, 16

female behavior service providers, and 1 male behavior service provider.

27

Demographic Questionnaire. The purpose of the demographic questionnaire (see Appendix C)

was to identify the children’s gender, age, primary language, ethnicity, race, and diagnosis.

Functional Analysis Screening Tool (FAST). The FAST (see Appendix D) was

completed prior to the WAIT for all children in Study II. The purpose of the FAST was to

identify a problem behavior to be targeted during the completion of the WAIT and the latency

FAs.

Waiting Assessment Interview Tool (WAIT). All informants were provided with an

electronic or physical copy of the WAIT and were asked to complete the questionnaire

independently. Moreover, the experimenters asked all informants if they had any questions

before or during their completion of the WAIT. If the informants had questions, the

experimenters answered all questions. In addition, all informants were asked to complete the

WAIT for the same target problem behavior. The target problem behavior was selected based on

the FAST. Questionnaires were provided in either English, Spanish, or in English with a

computer-based program used to translate information (i.e., MARTII). This program was only

used for Malik’s mom who spoke primarily Arabic.

Inter-Rater Reliability of the WAIT. To assess the inter-rater reliability of the WAIT,

some of the procedures used in the evaluation of the FAST (Iwata et al., 2013) were modified

and replicated. All informants who completed the WAIT were divided into pairs. Pairs consisted

of two caregivers or two behavior service providers, or both. Inter-rater reliability was calculated

by identifying the function selected to have the shortest latency of problem behavior by each

informant.

Modifying the procedures used by Iwata et al. (2013), the inter-rater reliability was

calculated by using the data gathered from both informants for each subject. The overall

28

reliability was assessed for each social function by identifying which function was selected as the

maintaining consequence of problem behavior by each informant. To do this, we selected the

shortest duration identified by the informants. If informants’ answers matched, suggesting that

waiting in one context (e.g., social positive for tangibles/edible) served as an establishing

operation to engage in problem behavior, agreement was scored. If informants’ answers did not

match, meaning each informant selected a different function, disagreement was scored.

Results.

Inter-Rater Reliability of the WAIT.

Caregivers. A WAIT was completed by at least one caregiver for all children in the

study. Overall, 6 pairs of caregivers completed the WAIT for the same child (see Table 1). From

those pairs, 4 identified the same shortest latency of problem behavior. Meaning, agreement was

found. For example, mom and dad identified that their child waited the shortest duration in the

tangible condition. No agreement was found for the remaining 2 pairs. Thus, there was

agreement for 67% of the caregivers.

Behavior Service Providers. A WAIT was only completed for children who received

ABA services. Therefore, a total of 8 pairs of behavior service providers completed the WAIT

(see Table 2). From those pairs, 4 pairs identified the same shortest latency of problem behavior.

Meaning, agreement was found. The remaining 4 pairs identified the shortest latency of problem

behavior for different functions. Meaning, no agreement was found. Thus, there was 50%

agreement and 50% disagreement for behavior service providers.

Discussion. Overall, 67% of caregivers and 50% of behavior service providers identified

a match for the function with the shortest latency. These results are preliminary but may suggest

that the WAIT may be an effective initial indirect tool to guide behavior service providers to

29

identify how long some children may wait under different circumstances to receive

reinforcement.

There are some limitations to these results. First, the number of informants who

completed the WAIT was different for caregivers and behavior service providers. Even though

11 children participated in Phase II of Study II, there were only six pairs of caregivers and eight

pairs of behavior service providers. It is possible that if we had used a controlled sample, more

informants would have been available to participate. However, the sample used is representative

of the individuals who may be available to complete questionnaires for children with

developmental disabilities. Moreover, from the caregivers who completed the WAIT, 11 mothers

participated. In contrast, only 5 fathers participated. It is possible that agreement would have

been higher if a different ratio of mothers to fathers would have participated. However, some

mothers who completed the WAIT reported to be single or divorced or reported that their

significant other was not available to complete the questionnaire. Another possible explanation

for the 33% of caregivers for which no agreement was scored could be the result of the

caregivers’ individual interactions with the children. It is likely that parents interact with their

children in different contexts and children may engage in discriminated responding when one

parent is present versus the other. For example, if mom is typically the person doing homework

with the child, the child may engage in problem behavior to escape from the tasks presented.

Therefore, if mom is asked to select under which circumstances the child engages in problem

behavior, mom is more likely to select an escape function. In contrast, dad is less likely to select

an escape function if he does not interact with the child during homework time.

In addition, only 50% correspondence was observed for the behavior service providers.

These results are more alarming because ideally behavior service providers work on the same

30

targets with the same child. However, there may be two explanations for this. First, the

difference in education and experience in behavior analysis may affect the results selected by the

providers. For example, a BCBA who completed the WAIT may have a better understanding of

functions of problem behavior than a newly certified RBT who may have only worked in the

field for several months. Second, the direct time working with a child may also affect these

results. For example, if two RBTs work with the same child but one of them has worked with the

child for a longer period of time or for more hours a week, their responses may vary as children

may behave differently with one RBT over another.

Overall, the inter-rater reliability data we obtained are not surprising with the number of

subjects who participated in the study. In general, these data are representative of previous

research using indirect assessments to identify circumstances under which problem behavior is

more likely to occur (e.g., Iwata et al., 2013). However, it is possible that if more informants are

selected per child and the study is conducted with more children, higher agreement percentages

may be identified, making the WAIT a more effective tool to be used to guide further assessment

and intervention of problem behavior.

Phase II – Direct Assessments

Method.

Subjects and Setting. The subjects for Study II Phase II consisted of 11 children who

engaged in problem behavior that warranted assessment. Informants in Study II Phase I

completed the questionnaires in relation to the children who participated in this phase.

David. 6-year-old Hispanic boy diagnosed with ASD. David engaged in tantrums which

was defined as vocalizations above conversation level that did not include requests for items or

people and crying with or without tears. His sessions were conducted at home in Florida.

31

Jinger. 10-year-old Caucasian girl diagnosed with ASD. Jinger engaged in aggression

which was defined as attempts or successes at biting, scratching, hair pulling, pinching, kicking,

or hitting others. Her sessions were conducted at home in Florida.

Issac. 9-year-old Hispanic boy who engaged in screaming which was defined as loud

vocalizations that did not include words. His sessions were conducted at home in Florida.

Blake. 3-year-old Caucasian boy diagnosed with ASD. Blake engaged in head-banging

which was defined as forceful contact of any part of his head with the floor, furniture, or another

person. His sessions were conducted at his ABA clinic in Florida.

Joaquin. 6-year-old Hispanic boy diagnosed with ASD. Joaquin engaged in aggression

which was defined as attempts or successes at kicking, hitting, hair pulling, throwing objects

toward others, or biting other individuals. His sessions were conducted at home in Florida.

Mahar. 8-year-old Middle-Eastern boy diagnosed with ASD. Mahar engaged in

screaming which was defined as loud vocalizations above conversational level that did not

include words. His sessions were conducted at his ABA clinic in Florida.

Sansa. 5-year-old Indian girl diagnosed with ASD and attention deficit hyperactivity

disorder. Sansa engaged in self-injury which was defined as pinching any part of her body. Her

sessions were conducted at home in Florida.

Alex. 7-year-old Caucasian boy diagnosed with ASD. Alex engaged in aggression which

was defined as attempts or successes at kicking, hitting, hair pulling, or biting others. His

sessions were conducted at his ABA clinic in Florida.

Javier. 10-year-old Hispanic boy diagnosed with ASD. Javier engaged in property

destruction which was defined as kicking or hitting the walls with opened hands or closed fists

with force. His sessions were conducted at his ABA clinic in Florida.

32

Allen. 5-year-old Caucasian boy diagnosed with ASD. Allen engaged in aggression

which was defined as hitting, kicking, or biting another person. His sessions were conducted at a

university-based clinic in Missouri.

Malik. 5-year-old Arabic boy diagnosed with ASD. Malik engaged in

elopement which was defined as attempting to or successfully leaving a designated area by

pulling up or down on a door handle. His sessions were conducted at a university-based clinic in

Missouri.

Preference Assessment. A multiple stimulus without replacement (DeLeon & Iwata,

1996) preference assessment was conducted with all subjects to identify preferred items to be

used during the different conditions of the latency FAs. Low-preferred items were used during

the attention and play sessions and high-preferred items were used during the tangible and play

sessions.

Response Measurement, Reliability, and Treatment Integrity. Researchers collected

data on the latency of problem behavior for all children during the latency FAs. Reliability of the

observation system was assessed by having a second researcher collect data on the latency of

problem behavior for at least 20% of all FA sessions for all subjects. To calculate interobserver

agreement (IOA) the shortest latency in seconds was divided over the longest latency in seconds

and multiplied by 100%. Treatment integrity (TI) data were collected to assess the researchers’

implementation of all the procedures during the FAs. Treatment integrity was collected for at

least 20% of all FA sessions for all subjects. To calculate treatment integrity, all correct steps

were divided over all steps in each condition of the latency FA. All results for IOA and TI may

be found in Table 3.

33

Latency Functional Analysis. The purpose of the latency FA was to identify the function

of the problem behavior and the average latency to engage in the problem behavior following the

procedures by Thomason-Sassi et al. (2011). For the purpose of this analysis, the lower the

number of min, the shorter the latency to problem behavior. Meaning, the problem behavior is

maintained by the function that sets the establishing operation for the individual to emit the

response in the shortest amount of time. The following conditions in the latency FA were

conducted using a multielement design: ignore (some subjects), attention, play, tangible, and

escape. The latency FA consisted of 5-min sessions with a 5-min inter-session interval. Sessions

were terminated after the first instance of problem behavior or after the five min elapsed.

Discriminative stimuli (SDs; Skinner, 1938) were used to enhance discrimination across

conditions (e.g., colored shirts; Conners et al., 2000).

Ignore. The ignore condition is a test condition used to identify if problem behavior is

maintained by automatic reinforcement. In this condition, the researcher and the subject were in

the same room and the researcher ignored all subject’s behavior. If the target problem behavior

happened during this condition, sessions were terminated after 1 min of the first instance of the

target problem behavior to avoid adventitious reinforcement. This condition was not conducted

with individuals who engaged in problem behavior that required the presence of another person

(e.g., aggression), as the problem behavior was unlikely maintained by automatic reinforcement.

Attention. The attention condition tests for problem behavior maintained by social

positive reinforcement in the form of attention. During this condition, the subject and the

researcher were in the session room together. At the beginning of the session, the researcher told

the subjects that he or she was going to be busy and gave the subject the opportunity to play with

a moderately preferred toy (e.g., “I’ll be working over here, you can play with your toy if you

34

want to”). Contingent on the first instance of the target problem behavior, the researcher

provided verbal attention (e.g., “No, do not hit yourself”), physical attention (e.g., hug), or both

types of attention, depending on what the informants reported, to the subjects and the session

was terminated. There were no programmed consequences for non-targeted behavior.

Play. The play condition was the control condition. All conditions were compared to the

control condition. In the play condition, the establishing operations to engage in behavior

maintained by social consequences were removed or reduced by providing non-contingent

attention, access to preferred items, and placing no demands on the subjects. In this condition,

the subject was provided with attention approximately every 30 s, free access to highly preferred

toys, and no demands. If the subjects emitted the target problem behavior, sessions were

terminated after 1 min of the response to avoid adventitious reinforcement. If no target behavior

occurred, sessions were terminated after 5 min. All non-targeted problem behavior was placed on

extinction.

Tangible. Rooker, Iwata, Harper, Fahmie, and Camp (2011) suggested that the use of

tangible conditions should be limited in FAs because of the possibility of producing a function

where there is not one (i.e., false positive result), especially for food items. Thus, the tangible

condition was only conducted with children whose informants reported that obtaining access to

items, or the removal of them, resulted in problem behavior. This condition tests for behavior

maintained by positive reinforcement in the form of access to preferred items (e.g., toys, food).

In this condition, the researcher and the subject were in the same room and the subject started the

session with access to highly preferred items. At the beginning of the session, the items were

removed by the researcher (e.g., “no more toys”). Contingent on the first instance of the target

35

problem behavior, the items were returned to the subject and the session was terminated. No

programmed consequences were provided for non-targeted problem behavior.

Escape. The escape condition tests for problem behavior maintained by negative

reinforcement in the form of task removal. In this condition, the researcher and subject were in

the same room. The session started when the researcher presented demands to the subject.

Demands were typically presented every 30 s using a three-step prompting procedure (Horner &

Keilitz, 1975). This procedure included verbal, model, and physical prompts. The demands, or

tasks, presented to each subject were based on the information provided by the informants. The

tasks that the informants identified as problematic (e.g., academic assignments, motor imitation,

daily-living activities) were used in this condition. Contingent on the target problem behavior,

tasks were discontinued, and work material was removed. Non-targeted problem behavior and

correct completion of the tasks were placed on extinction.

Results. Visual inspection was used to analyze the data from all latency FAs (see Figures

2 and 3). Researchers compared the latencies to respond in each test condition to the latency to

respond in the control condition. Shorter latencies in one or more test conditions suggest that

problem behavior is maintained by the function(s) tested in that condition.

David. Shorter latencies of tantrums were observed during the tangible (M = 56.60 s) and

escape (M = 175 s) conditions when compared to the play condition of the latency FA. These

results suggest that David’s tantrums were maintained by access to preferred items and escape

from demands.

Jinger. Shorter latencies of aggression were observed during the tangible (M = 4.60 s)

and escape (M = 162.60 s) conditions when compared to the play condition of the latency FA.

These results suggest that Jinger’s aggression was maintained by access to preferred items and

36

escape from demands.

Issac. Shorter latencies of screaming were observed during the tangible (M = 24.40 s) and

escape (M = 92.80) conditions when compared to the play condition of the latency FA. These

results suggest that Issac’s screaming was maintained by access to preferred items and escape

from demands.

Blake. Shorter latencies of head-banging were observed during the escape (M = 73.33 s)

condition when compared to the play condition of the latency FA. These results suggest that

Blake’s head-banging was maintained by escape from demands.

Joaquin. Shorter latencies of aggression were observed during the tangible (M = 147.75

s) condition when compared to the play condition of the latency FA. These results suggest that

Joaquin’s aggression was maintained by access to preferred items.

Mahar. Shorter latencies of screaming were observed during the tangible (M = 16.25 s)

and escape (M = 195.50 s) conditions when compared to the play condition of the latency FA.

These results suggest that Mahar’s screaming was maintained by access to preferred items and

escape from demands.

Sansa. Shorter latencies of self-injury were observed during the tangible (M = 117.17 s)

condition when compared to the play condition of the latency FA. These results suggest that

Sansa’s self-injury was maintained by access to preferred items.

Alex. Shorter latencies of aggression were observed during the tangible (M = 5.67 s) and

escape (M = 36.33 s) conditions when compared to the play condition of the latency FA. These

results suggest that Alex’s aggression was maintained by access to preferred items and escape

from demands.

Javier. Shorter latencies of property destruction were observed during the escape (M =

37

149 s) condition when compared to the play condition of the latency FA. These results suggest

that Javier’s property destruction was maintained by escape from demands.

Allen. Shorter latencies of aggression were observed during the tangible (M=141.43 s)

and escape (M= 137.00 s) conditions when compared to the play condition of the latency FA.

These results suggest that Allen’s aggression was maintained by access to preferred items and

escape from demands.

Malik. Shorter latencies of elopement were observed during the tangible (M= 103.40 s)

and attention (M= 134.50 s) conditions when compared to the play condition of the latency FA.

These results suggest that Malik’s elopement was maintained by access to preferred items and

attention from others.

General Method

Validity Assessments. To assess the validity of the WAIT two analyses were completed.

First, we evaluated correspondence between the functions identified by all informants to the

functions identified by the latency FA. To do this, we compared of the average latencies of

problem behavior identified by all informants and the average latency for each contingency (i.e.,

social positive for attention, social positive for access to tangibles, social negative for escape

from demands) identified as a function of problem behavior. Second, a simple linear regression

was completed to evaluate the extent to which the average latency identified by all informants

for each social function predicted the latency for problem behavior obtained in the last data point

for each social function identified in the latency FA.

Comparison of Functions. To compare the results from the WAITs completed by

informants to the latency FAs completed in controlled settings, two categories were used to

indicate a complete match: 1) the FA identified one function, the average of all WAITs identified

38

this function as having the shortest latency of problem behavior, 2) the FA identified two

functions, the average of all WAITs identified these functions as having the shortest latencies of

problem behavior. Moreover, one category was used to indicate a partial match 1) the FA

identified two functions, the average of all WAITs identified one of these functions as having the

shortest latency of problem behavior. If the FA identified a function and the average of all

WAITs did not identify this function as the function in which problem behavior had the shortest

latency, no match was identified.

Linear Regression. The purpose of the linear regression was to assess the extent to which

the results from the WAIT predict the results from the latency FA. Meaning, can the WAIT

identify the latency of problem behavior for the functions identified in the latency FA? The

research question for the linear regression was: Does the average score from the WAITs

completed by all informants predict the waiting duration of children with an ASD who engage in

problem behavior? The hypothesis for this analysis was that the average scores from the WAIT

predicted the current waiting duration of children with ASD who engage in problem behavior.

The null hypothesis for this analysis was that the average scores from the WAIT did not predict

the current waiting duration of children with ASD who engage in problem behavior. For the

purpose of this regression, the independent variable was the average number of seconds from the

WAIT identified by all informants for each social function. The dependent variable was the last

data point obtained from the latency FA for each social function. The number of seconds for

both, WAITs and latency FAs, were converted to numbers 1 thru 5. To do this, the session

duration (e.g., 300 s) was divided into 5 intervals of 60 s and all durations were rounded to the

nearest whole number. If the latency to engage in problem behavior was from 1 s to 60 s, the

number 1 was assigned. If the latency to engage in problem behavior was from 61 s to 120 s, the

39

number 2 was assigned. If the latency to engage in problem behavior was from 121 s to 180 s,

the number 3 was assigned. If the latency to engage in problem behavior was from 181 s to 240

s, the number 4 was assigned. Finally, if the latency to engage in problem behavior was from 241

s to 300 s, the number 5 was assigned. This was done to reduce the minimal differences in

seconds that may mask the results. For example, the difference between 25 s and 46 s may

suggest that the measures used cannot identify current waiting durations. However, the actual

difference between 25 s and 46 s is only 21 s which may be considered a negligible difference in

practical use. Moreover, because the latency FA is an assessment of problem behavior in which

contingencies are manipulated to obtain efficient results, the waiting durations of children may

decrease as the subjects contact those contingencies for several sessions. Thus, we may obtain

shorter latencies during the last sessions in which problem behavior occurs.

Results

Comparison of Functions. The results from the comparison of functions may be found

on Table 4. For 8 out of 11 children a match was identified. In addition, a partial match was

identified for 1 out of 11 children and no match was identified for 2 out of 11 children. These

results show that the WAIT identified the correct function or functions of problem behavior in

73% of the cases, a partial match is 9% of cases, and no match in 18% of cases.

Linear Regression. A linear bivariate regression was completed to determine if the

relationship between the average number of seconds from the WAIT identified by all informants

and the last data point obtained from the latency FA for each social function was statistically

significant. To do this, we used the last data point for the function(s) identified by the latency

FAs and the average latencies identified by all informants for the same function(s) identified by

the latency FAs. A total of 17 functions were used for this linear regression. A one-tailed t-test

40

resulted in a p-value of 0.0348 (alpha <0.05) and an R2 of 0.2641 (see Figure 4).

General Discussion

Latency FAs for all 11 children identified social functions. WAITs completed by all

informants also identified that the behavior problems were maintained by social contingencies.

After comparing the functions identified by both, the WAITs and Latency FAs, the results

showed that the WAITs identified the correct functions for 73% of the children. In addition, in

the two cases in which there was no match identified (Sansa and Javier), the WAITs identified

the correct function as the function with the second shortest latency. Although the purpose of the

WAIT is not to determine the function of problem behavior, it is interesting that the informants’

perceptions of latency to problem behavior in the presence of various EOs corresponded to the

identified functions of problem behavior in the latency FAs.

Furthermore, in the linear regression, the average of all WAITs predicted the duration

children waited before engaging in problem behavior to receive reinforcement. Overall, these

results suggest that the WAIT when completed by one or multiple informants (e.g., 1 to 4) may

identify under which circumstances children with developmental and intellectual disabilities

engage in problem behavior.

41

CHAPTER SEVEN

Study III

Purpose

The purpose of Study III was to use the WAIT completed by informants to systematically

individualize the initial component durations used during the schedule thinning procedure. A

secondary purpose was to use the informants’ preferred waiting times as final waiting targets for

all subjects.

Method

Subjects and Setting. Four children (Joaquin, Javier, Mahar, and Alex) who participated

in Study II Phase II also participated in Study III. Recall that all subjects had an ASD diagnosis

and engaged in socially-mediated problem behavior when waiting to receive reinforcement.

Sessions for Joaquin were conducted at home. Sessions for Javier, Mahar, and Alex were

conducted at their ABA clinics.

Response Measurement, Reliability, and Treatment Integrity. Data were collected

using a real-time behavioral data application on electronic devices (CounteeTM). Data were

collected on the rate of problem behavior and functional communication responses (FCRs),

prompts and compliance (Javier only), and duration of components during the multiple

schedules. A second observer collected data for at least 27% of all sessions for all subjects to

assess the reliability of the observation system. Sessions were divided into 10-s intervals and

partial-interval agreement was used to compare data collected by both observers. Partial-interval

agreement consisted of dividing the smaller frequency of behavior over the larger frequency of

42

behavior and multiplying the result by 100%. Zero instances of problem behavior by both

observers were scored as 100% IOA. The percentages for all intervals were averaged to calculate

IOA for the sessions (see Table 5 for results). Moreover, data on therapists’ behavior were

collected on at least 25% of sessions for all subjects to assess treatment integrity (e.g., researcher

reinforced all FCRs on a fixed ratio [FR] 1 schedule during FCT and the reinforcement

component of the multiple schedule, researcher placed all FCRs during the extinction component

of the multiple schedule on extinction, researcher ignored all problem behavior). See Table 6 for

results.

Design. A multiple baseline across subjects design was used during Study III.

Baseline. Noncontingent reinforcement (NCR)/ contingent reinforcement (CR) baselines

were completed following the latency FAs for all subjects. Sessions were 5 min and alternated

using a 1:2 ratio (1 NCR session to 2 CR sessions). Data were collected on rate of problem

behavior and functional communication responses (FCRs) for all subjects. During the NCR

sessions, subjects had free access to reinforcement (i.e., access to preferred items or breaks from

work) and problem behavior was ignored. During the CR sessions, subjects had access to

reinforcement contingent on the target problem behavior.

Functional Communication Training. Following NCR/CR baseline, an FCR was taught

to each subject to appropriately request for their social reinforcer as identified by the latency FA.

All FCRs were taught using FCT and were based on the behavioral repertoire of the subjects. For

all subjects, FCRs consisted of vocal responses. Functional communication training consisted of

5-min sessions in which a most-to-least prompting procedure was used. Furthermore, increased

prompt delays across sessions provided opportunities for independent responding (Touchette &

Howard, 1984). The initial session of the FCT for all subjects was implemented using a 0-s

43

prompt delay. The second session included a 1-s delay. Moreover, time delays increased across

sessions by 50% and were rounded to the nearest whole number.

All FCT sessions depended on the function of problem behavior identified by the latency

FA. For example, if problem behavior was maintained by access to toys, the beginning of the

sessions resembled the tangible condition in the FA with the addition of a rule provided by the

experimenter (e.g., “if you want to play with your toys, you can say ‘may I have my toys,

please’). After the rule was provided, the session began, and a researcher removed the toys and

provided an instruction (e.g., “no more toys”). Then, a second researcher implemented the

prompting procedure (e.g., a vocal prompt “may I have my toys, please”). Contingent on the

subject engaging in the FCR, with or without prompts, the toys were provided to the subjects for

30 s. Similarly, if problem behavior identified by the latency FA was maintained by escape from

demands, the sessions resembled the escape condition in the latency FA. During all FCT

sessions, all problem behavior, including the target problem behavior, was placed on extinction

and FCRs were reinforced on a fixed ratio (FR 1) schedule. Mastery criteria consisted of three

consecutive sessions with 100% independent responding and problem behavior below 80%

reduction from the last three baseline sessions.

Schedule Thinning. A multiple schedule procedure was conducted as the schedule

thinning method used to increase the time in which subjects had to wait to receive reinforcement.

The multiple schedule consisted of two components, reinforcement and extinction. During the

reinforcement component all FCRs were reinforced on a FR 1 schedule. During the extinction

component all FCRs were placed on extinction. Problem behavior was placed on extinction

during both components. In addition, the multiple schedules had schedule-correlated stimuli

(e.g., green laminated card during reinforcement, red laminated card during extinction) and the

44

addition of a rule stated by the experimenter at the beginning of each session (e.g., “when the

card is green, you may ask for a break by saying ‘break,’ when the card is red, you cannot ask for

a break, you have to work”; “when the card is green, you may ask for your toys by saying ‘may I

have my toys please’, when the card is red, you have to wait”).

The initial waiting durations (i.e., extinction component) for the multiple schedules

depended on the information provided by the informants who completed the WAIT. To identify

an initial waiting target, we used the shortest latency provided by one of the informants who

completed the WAIT in the setting in which we ran sessions. Meaning, if we completed sessions

at home, we compared the latencies from the caregivers who completed the WAIT and selected

the shortest latency as the initial waiting duration. For example, if a mother reported that her

child waited 5 s before engaging in problem behavior to receive reinforcement and a father

reported that the same child waited 20 s before engaging in problem behavior to receive

reinforcement, we used the mother’s reported duration. Therefore, we started the waiting

duration at 5 s. In addition, if an informant selected 0-s waiting time, we started the waiting

duration at 1 s.

All reinforcement components for all subjects started and remained at 60 s. During

reinforcement, contingent on FCRs, subjects had access to reinforcers for 30 s. In contrast, the

waiting durations for all subjects increased across sessions based on problem behavior. For

waiting durations to increase, problem behavior had to remain below the 80% reduction from the

last three sessions of baseline. We increased the waiting durations adapting the prompt delay

procedure implemented by Touchette and Howard (1984) for FCT. Thus, time delays increased

across sessions by 50% and were rounded to the nearest whole number until the final target

45

waiting duration was met. This procedure allowed for all subjects’ waiting durations to be

increased in an individualized but systematic format.

In addition, final target waiting durations were also individualized. We selected the final

waiting durations based on the goals identified on the WAIT by the informants. Similarly to the

selection of the initial waiting durations, we used the WAITs completed by the informants in the

setting in which we ran sessions. In this case, we selected the longest waiting time identified by

the informants. For example, if sessions were conducted at the clinic and WAITs were completed

by the BCBA and RBT who worked with the child, we selected our final goal to be the longest

waiting time selected by either the BCBA or RBT.

During the multiple schedules, all sessions were initially 5 min for all subjects. However,

sessions durations were increased to either 10 min, 11 min, 16 min, or a combination of any of

these to allow for individualized final waiting times. For example, if the final target duration for

a subject was 15 min, we ran sessions at 5 min until the addition of the extinction and

reinforcement components exceeded 5 min. At this time, we extended session durations to 10

and then 16 min. The final session durations were always 1 more min than the final waiting goals

to allow for the 60-s reinforcement component at the beginning of each session. Mastery criteria

for this phase consisted of three consecutive sessions with discriminated manding and problem

behavior below 80% reduction from the last three baseline sessions.

In order to calculate the response rate in each component to account for the differences in

duration of components and reinforcement time in the reinforcement component, two equations

were completed for each multiple schedule session. First, we divided the number of FCRs

emitted in the reinforcement component by the total duration of time in the reinforcement

component minus the reinforcement time divided by 60. Second, we divided the number of FCRs

46

emitted in the extinction component by the total duration of time in the extinction component

divided by 60.

Results

Figure 5 depicts the rate of problem behavior during NCR/CR baseline, FCT, and

multiple schedule phases for all subjects. Figure 6 depicts the rate of FCRs also emitted during

NCR/CR baseline, FCT, and multiple schedule phases for all subjects.

Baseline. During the NCR/CR baseline, all subjects engaged in higher rates of problem

behavior during the CR condition compared to the NCR condition. This suggests that problem

behavior was maintained by the social reinforcers identified in the latency FA. Moreover, zero

rates of FCRs were observed for all subjects.

Functional Communication Training. Following baseline, all subjects were trained to

request for their reinforcers using a vocal response. The vocal responses were “I want toys,

please” for Mahar, “break” for Javier, “I want toys” for Joaquin, and “May I have my toys,

please” for Alex. For Mahar and Joaquin, rates of FCRs increased immediately or after the first

session while rates of problem behavior decreased to zero or near zero levels. For Javier, rates of

FCRs remained at zero levels during the first three sessions but increased and remained at high

levels after the fourth session. Moreover, Javier’s rates of property destruction decreased

immediately when compared to baseline, however, we did not observe an 80% reduction from

the last three sessions of baseline until the last three sessions of the FCT. For Alex, rates of FCRs

remained at zero or near zero levels while rates of aggression increased when FCT was initially

implemented. This pattern of behavior remained for the first ten sessions of FCT. However,

during these sessions, Alex expressed that he did not want to engage in the response and he did

not want to play with the toys. Therefore, new stimuli, similar to the items that he had previously

47

engaged in problem behavior to access, were introduced and then included during all subsequent

sessions. This change resulted in an increase of FCRs and a reduction of problem behavior to

zero levels.

Schedule Thinning. When the multiple schedule procedure was implemented, problem

behavior remained at zero or near zero levels for three (Mahar, Joaquin, and Alex) of four

subjects. Moreover, FCRs remained at high levels during the reinforcement component and

decreased for all subjects during the extinction components as the sessions advanced.

Mahar. Sessions for Mahar were conducted at his ABA clinic. We used the results from

the WAITs completed by his behavior service providers. The shortest latency identified for

Mahar to wait for preferred items was 20 s. We started the multiple schedule using a 20-s waiting

period in extinction and 60 s in reinforcement. The longest waiting duration identified as a goal

in the WAIT was 15 min. Therefore, we used this as our final goal. Sessions started at 5 min.

During the 5-min multiple schedule, problem behavior remained at zero levels with the

exemption of two sessions. Discriminated responding was observed from the first sessions.

Functional communication responses remained at high levels during the reinforcement

component and near zero levels during the extinction component. After his extinction and

reinforcement durations exceed the 5-min session, we moved into 10-min session. Here, two

sessions were completed were problem behavior remained below the 80% reduction from the last

three sessions of baseline and FCRs were stable in the reinforcement component and near zero

levels in the extinction component. The final session duration was 16 min. This was done to

allow for the 15-min final waiting time and the 1-min reinforcement component. During the 16-

min sessions, problem behavior and FCRs in the extinction component remained near zero levels

while FCRs in the reinforcement component remained at high levels.

48

Javier. Sessions for Javier were conducted at his ABA clinic. We used the results from

the WAITs completed by his behavior service providers. The shortest latency identified for

Javier to wait was 20 s. We started the multiple schedule using a 20-s waiting period in

extinction and 60 s in reinforcement. The longest waiting duration identified as a goal in the

WAIT was 10 min. Therefore, we used this as our final goal. Sessions started at 5 min. During

the 5-min multiple schedule, problem behavior initially increased to baseline levels and FCRs in

both components did not occur during the first session. However, after the second session, we

observed high rates of FCRs during the reinforcement component and lower rates of FCRs in the

extinction component. Some variability in FCRs was observed until the last five sessions during

the 5-min sessions in which we observed consistent discrimination. Meaning, higher levels of

FCRs in reinforcement and lower levels of FCRs in extinction. As previously stated, we only

increased the extinction component by 50% when problem behavior was below the 80%

reduction from the last three sessions of baseline. Thus, we remained with the same component

durations across multiple sessions. For example, Javier experienced three sessions of 60-s

reinforcement and 20-s extinction, four sessions of 60-s reinforcement and 68-s extinction, and

five sessions of 60-s reinforcement and 102-s extinction. When the extinction component

reached 153 s, problem behavior decreased to near zero levels and extinction component

durations increased across sessions. Therefore, we increased the sessions to 10 min and then to

11 min to meet the final goal of 1-min reinforcement and 10-min extinction. During the 10-min

and 11-min sessions, discriminated responding was observed in all sessions. Higher levels of

FCRs occurred in the reinforcement component while lower levels of FCRs occurred in the

extinction component. Moreover, problem behavior remained near zero levels.

49

Joaquin. Sessions for Joaquin were conducted at home. We used the results from the

WAITs completed by his caregivers. The shortest latency identified for Joaquin was 0 s.

Therefore, we started the multiple schedule using a 1-s waiting period in extinction and 60 s in

reinforcement. The longest waiting duration identified as a goal in the WAIT by the caregivers

was 3 min. However, after talking to the caregivers and reviewing the WAITs completed by the

behavior service providers we decided to increase our final goal to what both behavior service

providers selected, 10 min. During the implementation of the multiple schedule using 5-min

sessions, problem behavior remained at zero or near zero rates. However, FCRs occurred at high

rates during both components of the multiple schedule with slightly higher levels of responding

during the reinforcement component. As sessions progressed, we continued to observe high rates

of FCRs during the reinforcement component and zero to near zero levels of FCR during the

extinction component. Therefore, we moved to 11-min sessions to account for the final goal of 1-

min reinforcement, 10-min extinction. Here, problem behavior remained at zero rates with the

exception of one session. In addition, FCRs remained at zero or near zero levels during the

extinction component and high levels during the reinforcement component.

Alex. Sessions for Alex were conducted at his ABA clinic. We used the results from the

WAITs completed by his behavior service providers. The shortest latency identified for Alex was

0 s. We started the multiple schedule using a 1-s waiting period in extinction and 60 s in

reinforcement. The longest waiting duration identified as a goal in the WAIT by the behavior

service providers was 15 min. Therefore, we used this as our final goal. During the

implementation of the multiple schedule using 5-min sessions, problem behavior was at zero or

near zero rates and discriminated responding was immediately observed during the sessions.

Thus, we moved to 10-min sessions. Here, we continued to observe zero levels of problem

50

behavior, high rates of FCRs during the reinforcement component, and zero to near zero levels of

FCRs during the reinforcement component. Finally, during the 16-min sessions, we continued to

observe low levels of problem behavior and high levels of FCRs in the reinforcement component

while FCRs in the extinction component remained at zero or near zero levels.

Discussion

Study III extends the current research on multiple schedules as a schedule thinning

method within the context of FCT for problem behavior maintained by social contingencies. In

addition, in this study, we evaluated a different procedure for selecting and increasing

component durations in the multiple schedules. To do this, we used the durations that the

caregivers or behavior service providers identified in the WAIT as the latency to problem

behavior as the initial waiting target for each subject. This method allowed for very short initial

extinction component duration targets (e.g., 1 s) while the reinforcement component duration

stayed consistent throughout all sessions (i.e., 60 s). Results suggest that acquisition of the

discriminated response for all subjects was obtained faster and for longer durations (e.g., 15 min)

from what has been previously reported in the research literature (e.g., Fisher et al., 2015). There

are two potential reasons for these results. First, it is possible that starting the extinction

components using individualized durations may have reduced potential extinction bursts

sometimes observed during the initial extinction components. Previous research has typically

selected arbitrary initial duration components during the multiple schedules (e.g., Betz et al.,

2013). Second, we used a different procedure from what has been used in previous literature to

increase the multiple schedule components. Rather than increasing the extinction components

arbitrarily, we adapted the prompt delay procedure evaluated by Touchette and Howard (1984)

within the context of the multiple schedule. We increased extinction component durations across

51

sessions by 50% and we rounded to the nearest whole number. In addition, we only used one

session with problem behavior below the 80% reduction from the last three sessions of baseline

as the criteria to increase the duration of extinction components. Previous research has typically

increased components using both, problem behavior and discriminated responding to determine

component increments (e.g., Campos et al., 2016) or have used at least two sessions with

problem behavior below a certain percentage reduction from baseline (e.g., 85%; Hanley et al.,

2001). These results may suggest that perhaps using both discriminated responding and problem

behavior to increase component durations may decrease the efficiency of the procedure as it may

take longer durations for discriminated responding to occur. It is possible that higher rates of

responding in the extinction component relative to the reinforcement component may occur

because, from the subject’s perspective, the FCR may be on a variable schedule of reinforcement

(e.g., variable interval [VI] or variable ratio [VR]). This may be especially true for individuals

whose problem behavior is maintained by negative reinforcement.

To date, the research that has evaluated the use of multiple schedules within the context

of FCT has not been effective at establishing discriminated responding without the use of other

treatment components in children who have problem behavior maintained by negative

reinforcement (e.g., response restriction; Fisher, Greer, Querim, & DeRosa, 2014; positive

reinforcement; Campos et al., 2017). However, in this study, we observed discriminated

responding without other treatment components for one subject who engaged in problem

behavior maintained by negative reinforcement, Javier. For Javier, high rates of problem

behavior were observed during FCT and the initial durations of the multiple schedule. Variable

rates of FCRs were observed in both components until the extinction components were

substantially longer than the reinforcement component. These results, although only available for

52

one subject, may suggest that extinction may take longer to be effective at reducing both problem

behavior and FCRs in the extinction component when behavior is maintained by negative

reinforcement. Future research should replicate these results with more individuals who engage

in problem behavior maintained by negative reinforcement.

Furthermore, there are some limitations that should be noted in this study. First, all four

subjects engaged in a vocal response to engage in the FCR. It is possible that we may have

obtained different results if we had used the procedures with individuals who communicated

using a different verbal response. However, the children who participated in the study, did not

have an age appropriate verbal repertoire even though they engaged in some vocal behavior. For

example, Javier only engaged in one-to-two-word sentences that were not always

understandable. His vocal repertoire consisted mostly of echoing other individuals in his

environment.

Second, for three of four subjects, sessions were conducted at their ABA clinics. It is

possible that the clinic itself, or some of the therapists present during some sessions, may have

had some stimulus control over the subjects’ learning behavior. This, could have increased the

likelihood of establishing discriminated responding more efficiently in all subjects. However,

similar results were observed for Joaquin for which we conducted sessions at home. In addition,

all subjects in this study were recommended to participate due to their severe problem behavior

in the clinic and at home.

Another limitation of this study was the different final goals selected by the informants

who completed the WAIT. In general, caregivers selected very short waiting durations (e.g., 3

min for Joaquin) while behavior service providers selected longer waiting durations (e.g., 15 min

for Mahar). To account for this, we chose the longest waiting duration identified by the

53

informant who completed the WAIT in the setting in which we ran sessions. However, for

Joaquin, sessions were conducted at home and caregivers selected three min of waiting. Rather

than following this goal, we chose the goal selected by the behavior service providers because

waiting three min did not seem to be age appropriate for Joaquin. In addition, in other settings

(e.g., school, ABA clinic) in which more children may need assistance from teachers or

therapists, waiting three min may not be sufficient for another individual to attend to Joaquin’s

request to receive access to a preferred item. Future research could evaluate waiting different

durations in different settings as identified by each informant or using an average waiting

duration based on all goals identified by all informants.

Finally, we did not evaluate the transfer of treatment effects with caregivers or behavior

service providers for the subjects in different settings. However, most therapists that worked with

Mahar, Javier, and Alex were present during most sessions of the study and at least one caregiver

was present in some sessions for Joaquin. Caregivers and therapists were not formally trained on

following all the procedures of the study. Rather, informal training was conducted when

therapists and caregivers asked questions about the procedures and after the study was

concluded. No data are available for this. Therefore, future research should evaluate the transfer

of treatment effects in different settings and with different caregivers.

54

CHAPTER EIGHT

CONCLUSION

Three studies were completed to create and evaluate a questionnaire to identify waiting

durations for subjects who engage in problem behavior to receive social positive or social

negative reinforcement. The first study consisted of the development of a questionnaire (i.e.,

WAIT). To complete this first step, 15 questions were submitted to an expert review of BCBA-

Ds who rated 93% of the questions as either relevant or very relevant. In addition, the experts

provided feedback about the overall questionnaire. All feedback was addressed by changing the

wording of questions, layout of the questionnaire, instructions, and removing all questions that

were rated below relevant or very relevant.

After developing the final version of the questionnaire, the WAIT was provided to

informants of children with an ASD who engage in problem behavior maintained by social

contingencies. The purpose of Study II Phase I was to assess the inter-rater reliability of the

questionnaire. To do this, the informants were divided into two categories: 1) caregivers and 2)

behavior service providers and the results provided by each set of informants were compared.

Here, we compared the results based on contexts in which problem behavior happens rather than

just latencies to problem behavior. This was done to account for potential similarities in latencies

but differences in contexts. For example, both caregivers may report that a child waits 20 s to

receive reinforcement. However, if these reports are for different contexts (e.g., engage in

problem behavior to receive attention vs. engage in problem behavior to escape task demands)

55

the results may not be as useful for treatment purposes. Overall, the results from the inter-rater

reliability showed that caregivers’ answers matched in 67% of the cases and behavior service

providers’ answers matched in 50% of cases.

To assess the validity of the questionnaire, Study II Phase II was completed. Here, the

average results provided by all informants were compared to the last data point in the latency

FAs for each social function identified by the latency FAs completed in controlled settings. In

addition, a linear regression was completed to identify if the WAIT predicted the duration

children waited to receive reinforcement before engaging in problem behavior. The results from

the first analysis showed that in 73% of the cases, the WAIT identified the correct function or

functions that the latency FA identified. Moreover, the linear regression identified that the WAIT

predicted the duration children wait before engaging in problem behavior to receive

reinforcement.

Finally, Study III was used to further validate the use of the WAIT by systematically

individualizing the component durations used during a schedule thinning procedure. To do this,

we used the current waiting durations and the final waiting target durations identified by the

informants. Results showed that using individualized beginning and ending targets was

successful in obtaining discriminated responding while problem behavior remained at zero or

near zero levels for all subjects.

In general, the current set of studies extend the existing research on indirect assessments

and provide an interview tool that may be used by researchers and practitioners. This tool may be

useful when working with individuals with an ASD who engage in problem behavior when

waiting to receive reinforcement. Overall, we do not expect the WAIT to be the only assessment

completed when working with children with problem behavior. Instead, we expect the WAIT to

56

be used as an initial evaluation that could help develop further direct assessments and treatments.

In addition, these studies expand the current research on schedule thinning by evaluating

individualized initial and final duration components during multiple schedules which may

increase the durability of treatment effects and decrease the recurrence of problem behavior.

57

Table 1. Inter-rater Reliability of WAIT (caregivers).

Subjects Caregiver 1 Caregiver 2 Results

David Tangible Tangible Agreement

Joaquin Tangible Tangible Agreement

Mahar Escape Escape Agreement

Sansa Escape Escape Agreement

Jinger Escape Tangible Disagreement

Issac Escape Tangible Disagreement

Blake Tangible - -

Alex Tangible - -

Javier Escape - -

Malik Tangible - -

Allen Tangible - -

58

Table 2. Inter-Rater Reliability of WAIT (behavior service providers).

Subjects Behavior Service Provider 1

Behavior Service Provider 2 Results

Issac Tangible and Escape Tangible and Escape Agreement

Blake Escape Escape Agreement

Mahar Tangible Tangible Agreement

Alex Tangible Tangible Agreement

Jinger Escape Tangible and Attention Disagreement

Joaquin Escape Tangible Disagreement

Sansa Tangible Escape Disagreement

Javier Attention Tangible Disagreement

David - - -

Malik - - -

Allen - - -

59

Table 3. Interobserver Agreement (IOA) and Treatment Integrity (TI) for Latency Functional Analyses.

IOA % Sessions

IOA % Mean

IOA % Range

TI % Sessions

TI % Mean

TI % Range

David 28 99.77 98.38 - 100 40 100 -

Jinger 45 88.98 83.33 - 100 40 98.21 85.71 - 100

Issac 36 99.96 99.60 - 100 20 100 -

Blake 20 100 - 33 96.00 80.00 - 100

Joaquin 44 99.89 99.20 - 100 35 100 -

Mahar 25 96.03 83.87 - 100 33 100 -

Sansa 30 100 - 25 100 -

Alex 33 93.45 75.00 - 100 67 100 -

Javier 33 100 - 67 100 -

Allen 37 94.66 42.85 - 100

Malik 36 97.90 90.91 - 100

60

Table 4. Comparison of Functions for WAITs and Latency Functional Analyses.

Subjects Latency FAs All WAITs Results

David 1. Tangible 2. Escape

1. Tangible 2. Escape 3. Attention

Match

Jinger 1. Tangible 2. Escape

1. Tangible 2. Escape 3. Attention

Match

Issac 1. Tangible 2. Escape

1. Escape 2. Tangible 3. Attention

Match

Blake 1. Escape

1. Escape 2. Tangible 3. Attention

Match

Joaquin 1. Tangible

1. Tangible 2. Escape 3. Attention

Match

Mahar 1. Tangible 2. Escape

1. Tangible 2. Escape

Match

Alex 1. Tangible 2. Escape

1. Tangible 2. Escape

Match

Malik 1. Tangible 2. Attention

1. Attention 2. Tangible

Match

Allen

1. Tangible 2. Escape

1. Tangible 2. Attention 3. Escape

Partial Match

Sansa 1. Tangible

1. Escape 2. Tangible 3. Attention

No Match

Javier 1. Escape

1. Tangible 2. Escape 3. Attention

No Match

61

Table 5. Interobserver Agreement (IOA) for Study III.

Subjects % Sessions

Mean and Range % Problem Behavior

Mean and Range % FCRs SD

Mean and Range % FCRs SD

Mahar 33 99.16 (96.66, 100) 99.05 (96.66, 100) 99.67 (98.33, 100)

Javier 27 95.53 (89.30, 100) 99.17 (97.50, 100) 92.30 (84.16, 95.83)

Joaquin 33 99.58 (98.33, 100) 99.58 (96.66, 100) 98.53 (96.21, 100)

Alex 32 97.91 (88.33, 100) 99.25 (96.66, 100) 97.78 (93.33, 100)

62

Table 6. Treatment Integrity (TI) for Study III.

Subjects % Sessions

TI % Mean

TI % Range

Mahar 25 100 -

Javier 30 100 -

Joaquin 33 99.50 98.36 - 100

Alex 33 100 -

63

Figure 1. Expert review responses to initial WAIT questions. The rating scale is shown on the x-axis and consisted of Very Relevant, Relevant, Slightly Relevant, or Not Relevant. The percentage of expert review responses to all questions is shown on the y-axis.

Very Rele

vant

Releva

nt

Slightl

y Rele

vant

Not Rele

vant

0

20

40

60

80

100

EXPE

RT R

EVIE

W R

ESPO

NSE

S T

O A

LL Q

UES

TIO

NS

(%)

64

Figure 2. Latency functional analyses for six subjects. On the top panel David and Jinger, on the middle panel Issac and Blake, and on the bottom panel Joaquin and Mahar. Sessions are on the x-axis and latency of problem behavior on the y-axis. Closed symbols represent the ignore condition, open circles represent the play condition, open squares represent the attention condition, closed squares represent the tangible condition, and closed triangles represent the escape condition.

5 10 15 20 250

100

200

300

SESSION

LATE

NCY

OF

TAN

TRU

M (s

)

David

Attention

Escape

Play

Tangible

Ignore

5 10 15 20 250

100

200

300

SESSION

LATE

NCY

OF

SCRE

AM

ING

(s)

Isaac

Attention

Escape

Play

Tangible

Ignore

5 10 15 200

100

200

300

SESSION

LATE

NCY

OF

AG

GRE

SSIO

N (s

)

Joaquin

EscapePlay

Tangible

Attention

5 10 15 200

100

200

300

SESSION

LATE

NCY

OF

AG

GRE

SSIO

N (s

)

Jinger

AttentionEscape

Play

Tangible

5 10 150

100

200

300

SESSION

LATE

NCY

OF

HEA

D-B

AN

GIN

G (s

)

Blake

Ignore

Escape

Play Tangible Attention

5 10 15 200

100

200

300

SESSION

LATE

NCY

OF

SCRE

AM

ING

(s)

Mahar

Ignore

Escape

Play

Tangible

Attention

65

Figure 3. Latency functional analyses for five subjects. On the top panel Sansa and Alex, on the middle panel Javier and Allen, and on the bottom panel Malik. Sessions are on the x-axis and latency of problem behavior on the y-axis. Closed symbols represent the ignore condition, open circles represent the play condition, open squares represent the attention condition, closed squares represent the tangible condition, and closed triangles represent the escape condition.

5 10 15 20 25 300

100

200

300

SESSION

LATE

NC

Y O

F SE

LF-IN

JURY

(s)

Sansa

Attention

EscapePlay

Tangible

Ignore

2 4 6 8 10 120

100

200

300

SESSION

LATE

NC

Y O

F A

GG

RES

SIO

N (s

)

Alex

Attention

Escape

Play

Tangible

5 10 150

100

200

300

SESSION

LATE

NC

Y O

F PR

OPE

RTY

D

ESTR

UC

TIO

N (s

)

Javier

Attention

Escape

Play TangibleIgnore

5 10 15 20 25 30 350

100

200

300

SESSION

LATE

NC

Y O

F A

GG

RES

SIO

N (s

)

Allen

Attention

Escape

Play

Tangible

2 7 12 17 220

100

200

300

SESSION

LATE

NC

Y O

F EL

OPE

MEN

T (s

)

Malik

Attention

Escape Play

Tangible

66

Figure 4. Linear regression for function(s) of problem behavior. Informants endorsed latencies (min) on the x-axis and last data point in latency FAs (min) on the y-axis.

0 1 2 3 4 50

1

2

3

4

5

INFORMANTS ENDORSED LATENCIES (MIN)

LAST

DAT

A PO

INT

IN L

ATEN

CY F

As (

MIN

)

FUNCTION(S) OF PROBLEM BEHAVIOR

67

Figure 5. Rate of problem behavior for four subjects during baseline, functional communication training, and multiple schedules. Open circles represent rate of problem behavior in the NCR condition. Closed circles represent rate of problem behavior in the CR condition, FCT, and all phases of the multiple schedules. Mahar’s data are shown on the first panel, Javier’s data are shown on the second panel, Joaquin’s data are shown on the third panel, and Alex’s data are shown on the fourth panel.

0

2

4

6

8

10

PRO

PERT

Y D

ESTR

UC

TIO

N (R

PM)

Javier

0

1

2

SCR

EAM

ING

(RPM

)

CR

NCR

Mahar

BL FCTMS

5 minMS

10 minMS

16 min

MS 11 min

5 10 15 20 25 30 35 40 45 500

1

2

3

4

SESSION

AG

GR

ESSI

ON

(RPM

)

Alex

0

1

2

AG

GR

ESSI

ON

(RPM

)

Joaquin

MS 16 min

MS 11 min

68

Figure 6. Rate of functional communication responses for all four subjects during baseline, functional communication training, and multiple schedules. Open circles represent rate of FCRs in the NCR condition in BL and in the reinforcement (SD) component during the multiple schedules. Closed circles represent rates of FCRs in the CR condition in BL, FCT, and in the extinction (SD) component during the multiple schedules. Mahar’s data are shown on the first panel, Javier’s data are shown on the second panel, Joaquin’s data are shown on the third panel, and Alex’s data are shown on the fourth panel.

0 5 10 15 20 25 30 35 40 45 500

20

40

60

80

SESSION

Alex

0

10

20

30

40

Joaquin

0

10

20

30

40

CRNCR

Mahar

SΔ

SD

0

10

20

30

40

50

Javier

BL FCTMS

5 minMS

10 minMS

16 min

MS 11 min

FUN

CTIO

NA

L CO

MM

UN

ICAT

ION

RES

PON

SES

(RPM

)

MS 16 min

MS 11 min

69

REFERENCES

American Psychiatric Association. (2013). Diagnostic and statistical manual of

mental disorders (5th ed.). Washington, DC. Beavers, G. A., Iwata, B. A., & Lerman, D. C. (2013). Thirty years of research on the

functional analysis of problem behavior. Journal of Applied Behavior Analysis, 46, 1–21. doi:10.1002/jaba.30

Betz, A. M., Fisher, W. W., Roane, H. S., Mintz, J. C., & Owen, T. M. (2013). A

component analysis of schedule thinning during functional communication

training. Journal of Applied Behavior Analysis, 46, 219–241. doi:10.1001/jaba.23

Bloom, S. E., Iwata, B. A., Fritz, J. N., Roscoe, E. M., & Carreau, A. B.

(2011). Classroom application of a trial-based functional analysis. Journal of Applied

Behavior Analysis, 44, 19–31. doi:10.1901/jaba.2011.44-19

Campos, C., Leon, Y., Sleiman, A., & Urcuyo, B. (2016). Further evaluation of the use of

multiple schedules for behavior maintained by negative reinforcement. Behavior Modification, 41, 269–285. doi:10.1177/0145445516670838

Carr, E. G. & Durand, V. M. (1985). Reducing behavior problems through functional

communication training. Journal of Applied Behavior Analysis, 18, 111–126.

doi:10.1901/jaba.1985.18-111

70

Carr, J. E. & Fox, E. J. (2009). Using video technology to disseminate behavioral

procedures: A review of functional analysis: A guide for understanding challenging behavior (dvd). Journal of Applied Behavior Analysis, 42, 919–923. doi:10.1901/jaba.2009.42-919

Centers for Disease Control and Prevention (2016, August). U.S. Census Bureau. (2016, May). Facts about developmental disabilities. U.S. Department of Health and Human Services. Retrieved from https://www.cdc.gov/ncbddd/developmentaldisabilities/facts.html

Christensen, D. L., Baio, J., Braun, K. V., Bilder, D., Charles, J., Constantino, J. N., …

Yeargin-Allsopp, M. (2016). Prevalence and characteristics of autism spectrum disorder among children aged 8 years–autism and developmental disabilities monitoring network, 11 sites, United States, 2012. Surveillance Summaries, 65, 1–23. doi:10.15585/mmwr.ss6503a1

Conners, J., Iwata, B. A., Kahng, S., Hanley, G. P., Worsdell, A. S., & Thompson, R. H.

(2000). Differential responding in the presence and absence of discriminative stimuli during multielement functional analyses. Journal of Applied Behavior Analysis, 33, 299–308. doi:10.1901/jaba.2000.33–299

Dehdari, T., Rahimi, T., Aryaeian, N., Gohari, M. R., & Esfeh, J. M. (2014). Developing

and testing a measurement tool for assessing predictors of breakfast consumption based on a health promotion model. Journal of Nutrition Education and Behavior, 46, 250–258. doi:10.1016/j.jneb.2013.12.007

DeLeon, I. G., & Iwata, B. A. (1996). Evaluation of a multiple-stimulus presentation format for

assessing reinforcer preferences. Journal of Applied Behavior Analysis, 29, 519-533.

doi:10.1901/jaba.1996.29-519

71

Dixon, M. R. & Cummings, A. (2001). Self-control in children with autism: response

allocation during delays to reinforcement. Journal of Applied Behavior Analysis, 34, 491–495. doi:10.1901/jaba.2001.34–491

Dominick, K. C., Davis, N. O., Lainhart, J., Tager-Flusberg, H., & Folstein, S. (2007).

Atypical behaviors in children with autism and children with a history of language impairment. Research in Developmental Disabilities, 28, 145–162. doi: 10.1016/j.ridd.2006.02.003

Durand, V. M., & Crimmins, D. B. (1988). Identifying the variables maintaining self-

injurious behavior. Journal of Autism and Developmental Disorders, 18, 99–117.

doi:10.1007/BF02211821

Fisher, W. W., Greer, B. D., Fuhrman, A. M., & Querim, A. C. (2015). Using multiple

schedules during functional communication training to promote rapid transfer of

treatment effects. Journal of Applied Behavior Analysis, 48, 713–733.

doi:10.1002/jaba.254

Fisher, W. W., Greer, B. D., Querim, A. C., & DeRosa, N. (2014). Decreasing excessive

functional communication responses while treating destructive behavior using response

restriction. Research in Developmental Disabilities, 35, 2614-2623.

doi:10.1016/j.ridd.2014.06.024

Fisher, W. W., Piazza, C. C., Bowman, L. G., & Amari, A. (1996). Integrating caregiver

report with a systematic choice assessment. American Journal on Mental

Retardation, 101,15–25.

72

Fisher, W., Piazza, C., Cataldo, M., Harrell, R., Jefferson, G., & Conner, R.

(1993). Functional communication training with and without

extinction and punishment. Journal of Applied Behavior Analysis,

26, 23–36. doi:10.1901/jaba.1993.26-23

Fisher, W. W., Thompson, R. H., Hagopian, L., Bowman, L. G., & Krug,

A. (2000). Facilitating tolerance of delayed reinforcement during

functional communication training. Behavior Modification, 24, 3–29.

doi:10.1177/0145445500241001

Green, C. W., Reid, D. H., Canipe, V. S., & Gardner, S. (1991). A comprehensive

evaluation of reinforcer identification processes for persons with profound multiple handicaps. Journal of Applied Behavior Analysis, 24, 537–552. doi:10.1901/jaba.1991.24–537

Gresham, F. M., & Elliott, S. N. (2008). Social Skills Improvement System. Minneapolis,

MN: Pearson Assessments.

Hagopian, L. P., Boelter, E. W., & Jarmolowicz, D. P. (2011). Reinforcement

Schedule thinning following functional communication training: Review and

recommendations. Behavior Analysis in Practice, 4, 4–16.

doi:10.1016/j.ridd.2011.05.009

Hanley, G. P., Heal, N. A., Tiger, J. H., & Ingvarsson, E. T. (2007). Evaluation of a

classwide teaching program for developing preschool life skills. Journal of Applied Behavior Analysis, 40, 277–300. doi:10.1901/jaba.2007.57–06

73

Hanley, G. P., Iwata, B. A., & Thompson, R. H. (2001). Reinforcement schedule thinning

following treatment with functional communication training. Journal of Applied

Behavior Analysis, 34, 17–38. doi:10.1901/jaba.2001.34–17

Horner, R. H. & Day, H. (1991). The effects of response efficiency on functionally

equivalent competing behaviors. Journal of Applied Behavior Analysis, 24, 719–

732. doi:10.1901/jaba.1991.24-719

Horner, R. D. & Keilitz, I. (1975). Training mentally retarded adolescents to brush their

teeth. Journal of Applied Behavior Analysis, 8, 301–309.

doi:10.1901/jaba.1975.8-301

Iwata, B. A., DeLeon, I. G., & Roscoe, E. M. (2013). Reliability and validity of the

functional analysis screening tool. Journal of Applied Behavior Analysis, 46, 271– 284. doi:10.1002/jaba.31

Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1982/1994).

Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27,

197–209. doi:10.1901/jaba.1994.27-197

Jin, C. S., Hanley, G. P., & Beaulieu, L. (2013). An individualized and comprehensive

approach to treating sleep problems in young children. Journal of Applied Behavior Analysis, 46, 161–180. doi:10.1002/jaba.16

Kuhn, D. E., Chirighin, A. E., & Zelenka, K. (2010). Discriminated functional

communication: A procedural extension of functional communication training. Journal of Applied Behavior Analysis, 43, 249–264. doi:10.1901/jaba.2010.43– 249

74

Lalli, J. S., Casey, S., & Kates, K. (1995). Reducing escape behavior and increasing task

completion with functional communication training, extinction and response chaining. Journal of Applied Behavior Analysis, 28, 261–268. doi:10.1901/jaba.1995.28-261

Laraway, S., Snycerski, S., Michael, J., & Poling, A. (2003). Motivating operations and

terms to describe them: Some further refinements. Journal of Applied Behavior Analysis, 36, 407–414. doi:10.1901/jaba.2003.36-407

Leon, Y., Hausman, N. L., Kahng, S., & Becraft, J. L. (2010). Further examination of discriminated functional communication. Journal of Applied Behavior Analysis, 43, 525–530. doi:10.1901/jaba.2010.43–525

Lovaas, O. I. (1993). The development of a treatment-research project for

developmentally disabled and autistic children. Journal of Applied Behavior Analysis, 26, 617–630. doi:10.1901/jaba.1993.26-617

Matson, J. L., Rotatori, A. F., & Helsel, W. J. (1983). Development of a rating scale to

measure social skills in children: The Matson Evaluation of Social Skills with

Youngsters (MESSY). Behavior Research and Therapy, 21, 335–340.

Matson, J. L., & Vollmer, T. R. (1995). User’s guide: Questions About Behavioral

Function (QABF). Baton Rouge, LA: Scientific Publishers. Newquist, M. H., Dozier, C. L., & Neidert, P. L. (2012). A comparison of the effects of

brief rules, a timer, and preferred toys on self-control. Journal of Applied Behavior Analysis, 45, 497–509. doi:10.1901/jaba.2012.45–497

75

Northup, J., Wacker, D., Sasso, G., Steege, M., Cigrand, K., Cook, J., & DeRaad, A.

(1991). A brief functional analysis of aggressive and alternative behavior in an outclinic setting. Journal of Applied Behavior Analysis, 24, 509–522. doi:10.1901/jaba.1991.24-509

Passage, M., Tincani, M., & Hantula, D. A. (2012). Teaching self-control with

qualitatively different reinforcers. Journal of Applied Behavior Analysis, 45, 853– 857. doi:10.1901/jaba.2012.45–853

Perez, C., Sawmiller D., & Tan, J. (2016). The role of heparan sulfate deficiency in

autistic phenotype: Potential involvement of Slit/Robo/srGAPs-mediated dendritic

spine formation. Neural Development, 11, 11. doi:10.1186/s13064-016-0066-x

Querim, A. C., Iwata, B. A., Roscoe, E. M., Schlichenmeyer, K. J., Ortega, J. V., & Hurl,

K. E. (2013). Functional analysis screening for problem behavior maintained by

automatic reinforcement. Journal of Applied Behavior Analysis, 46, 47–60.

doi:10.1002/jaba.26

Reimers, T. M., Wacker, D., & Koeppl, G. (1987). Acceptability of behavioral interventions: A review of the literature. School Psychology Review, 212-227. Rooker, G. W., Iwata, B. A., Harper, J. M., Fahmie, T. A., & Camp, E. M. (2011). False-

positive tangible outcomes of functional analyses. Journal of Applied Behavior Analysis,

44, 737–745. doi:10.1901/ jaba.2011.44-737

Rooker, G. W., Jessel, J., Kurtz, P. F., & Hagopian, L. P. (2013). Functional

communication training with and without alternative reinforcement and

punishment: An analysis of 58 applications. Journal of Applied Behavior

Analysis, 46, 708–722. doi:10.1002/jaba.76

76

Schweitzer, J. B., & Sulzer Azaroff, B. (1988). Self-control: Teaching tolerance for delay

in impulsive children. Journal of the Experimental Analysis of Behavior, 50,

173–186. doi:10.1901/jeab.1988.50-173

Skinner, B. F. (1938). The behavior of organisms: An experimental

analysis. New York: Appleton-Century-Crofts.

Smith, R. G., & Churchill, R. M. (2002). Identification of environmental

determinants of behavior disorders through functional analysis of precursor

behaviors. Journal of Applied Behavior Analysis, 35, 125–136.

doi:10.1901/jaba.2002.35-125v

Stokes, T. F. & Baer, D. M. (1977). An implicit technology of generalization. Journal of Applied Behavior Analysis, 10, 349-367.

Strain, P. S., Schwartz, I. S., & Barton, E. E. (2011). Providing interventions for young

children with autism spectrum disorders what we still need to accomplish. Journal of

Early Intervention, 33, 321–332.

Thomason-Sassi, J. L., Iwata, B. A., Neidert, P. L., & Roscoe, E. M. (2011). Response

latency as an index of response strength during functional analyses of problem

behavior. Journal of Applied Behavior Analysis, 44, 51–67. doi:10.1901/jaba.2011.44-51

Tiger, J. H., Hanley, G. P., & Bruzek, J. (2008). Functional Communication Training: A

Review and Practical Guide. Behavior Analysis in Practice, 1, 16–23. Touchette, P. E., & Howard, J. S. (1984). Errorless learning: Reinforcement

contingencies and stimulus control transfer in delayed prompting. Journal

of Applied Behavior Analysis, 17, 175–188. doi:10.1901/jaba.1984.17-175

77

Vollmer, T. R., Borrero, J. C., Lalli, J. S., & Daniel, D. (1999). Evaluating self-control

and impulsivity in children with severe behavior disorders. Journal of Applied Behavior Analysis, 32, 451–466. doi:10.1901/jaba.1999.32–451

Zarcone, J. R., Rodgers, T. A., Iwata, B. A., Rourke, D., & Dorsey, M. F. (1991).

Reliability analysis of the Motivation Assessment Scale: A failure to replicate. Research in Developmental Disabilities, 12, 349–360.

78

APPENDIX A:

WAITING ASSESSMENT INTERVIEW TOOL (ENGLISH)

79

APPENDIX B:

WAITING ASSESSMENT INTERVIEW TOOL (SPANISH)

80

APPENDIX C:

DEMOGRAPHIC QUESTIONNAIRE

**Please answer the questions by circling the answer that best applies to you.**

Questions About the Participant

1. What is the participant’s gender?

Male Female

2. What is the participant’s age?

__________

3. What is the participant’s primary language?

English Spanish French Other ______________

4. What is the participant’s ethnicity?

Hispanic or Latino Not Hispanic or Latino Other: _______________

5. What’s the participant’s race?

American Indian/Alaskan Native Asian Native Hawaiian/Pacific Islander Black or African American White Other: _________________

6. What is the participant’s diagnosis?

81

APPENDIX D:

FUNCTIONAL ANALYSIS SCREENING TOOL (FAST)

82

APPENDIX E:

IRB APPROVAL LETTER

10/24/2017

Claudia Campos ABA-Applied Behavior Analysis 4202 E Fowler Ave Tampa, FL 33620

RE: Expedited Approval for Initial Review

IRB#: Pro00030383

Title: Evaluation of the waiting assessment interview tool (WAIT) and individualized waiting durations in signaled reinforcement

Study Approval Period: 10/23/2017 to 10/23/2018

Dear Ms. Campos:

On 10/23/2017, the Institutional Review Board (IRB) reviewed and APPROVED the above application and all documents contained within, including those outlined below.

Approved Item(s): Protocol Document(s): WAIT and Signal Reinforcement Protocol_IRB_Version 1_10.1.17.docx

Consent/Assent Document(s)*:

WAIT and signal reinforcement_Parenand Parental Consent_Version 1_10.10.17 No HIPAA.docx.pdf WAIT and signal reinforcement_Parent and Parental Consent_Version 1_8.30.17.docx.pdf ***WAIT and Signaled Reinforcement_Verbal Consent_10.1.17 .docx ***WAIT and Signaled Reinforcement_Verbal Consent with HIPAA Auth.docx

83

Please use only the official IRB stamped informed consent/assent document(s) found under the "Attachments" tab. Please note, these consent/assent documents are valid until the consent document is amended and approved.***Consent forms with waiver are not stamped.

It was the determination of the IRB that your study qualified for expedited review which includes activities that (1) present no more than minimal risk to human subjects, and (2) involve only procedures listed in one or more of the categories outlined below. The IRB may review research through the expedited review procedure authorized by 45CFR46.110 and 21 CFR 56.110. The research proposed in this study is categorized under the following expedited review category:

(6) Collection of data from voice, video, digital, or image recordings made for research purposes.

(7) Research on individual or group characteristics or behavior (including, but not limited to, research on perception, cognition, motivation, identity, language, communication, cultural beliefs or practices, and social behavior) or research employing survey, interview, oral history, focus group, program evaluation, human factors evaluation, or quality assurance methodologies.

Your study qualifies for a waiver of the requirements for the documentation of informed consent for screening/recruitment purposes as outlined in the federal regulations at 45CFR46.117(c) which states that an IRB may waive the requirement for the investigator to obtain a signed consent form for some or all subjects if it finds either: (1) That the only record linking the subject and the research would be the consent document and the principal risk would be potential harm resulting from a breach of confidentiality. Each subject will be asked whether the subject wants documentation linking the subject with the research, and the subject's wishes will govern; or (2) That the research presents no more than minimal risk of harm to subjects and involves no procedures for which written consent is normally required outside of the research context.

Your study qualifies for a waiver of the requirement for signed authorization as outlined in the HIPAA Privacy Rule regulations at 45CFR164.512(i) which states that an IRB may approve a waiver or alteration of the authorization requirement provided that the following criteria are met (1) the PHI use or disclosure involves no more than a minimal risk to the privacy of individuals; (2) the research could not practicably be conducted without the requested waiver or alteration; and (3) the research could not practicably be conducted without access to and use of the PHI. An alteration of HIPAA Authorization is granted for participants recruited from Engage Behavioral Health or the Silver Center at USF whose parents will provide Authorization verbally as part of the screening/recruitment process. This alteration exempts the study team from the Privacy Rule's requirement that Authorizations obtained during screening be signed and dated. Written Authorization will be obtained from the parents of those participants from these centers who meet inclusion criteria and decide to participate in the research as part of the informed consent process. As the PI is not part of the USF Covered Entity (CE), Authorization is not required from participants who are recruited from non-CEs.

This study involving child participants falls under the minimal risk category 45 CFR 46.404: Research not involving greater than minimal risk.

84

As the principal investigator of this study, it is your responsibility to conduct this study in accordance with IRB policies and procedures and as approved by the IRB. Any changes to the approved research must be submitted to the IRB for review and approval via an amendment. Additionally, all unanticipated problems must be reported to the USF IRB within five (5) calendar days.

We appreciate your dedication to the ethical conduct of human subject research at the University of South Florida and your continued commitment to human research protections. If you have any questions regarding this matter, please call 813-974-5638.

Sincerely,

Kristen Salomon, Ph.D., Vice Chairperson

USF Institutional Review Board